alirezarezvani/claude-skills
227 skills172k installs4M starsGitHub
Install
npx skills add https://github.com/alirezarezvani/claude-skillsSkills in this repo
1Marketing SkillsMarketing Skills Division is a journey-wide marketing operating system for solo and indie builders who ship with Claude Code, Codex, Cursor, OpenClaw, and similar agents. Instead of one generic copywriter prompt, you load marketing-ops as a router that sends you to the right specialist among forty-two skills grouped into seven pods: content strategy through social, traditional and AI SEO, conversion-focused page and funnel work, paid and email channels, growth experiments, intelligence, and sales enablement. A marketing-context foundation layer keeps brand voice, audience, and goals consistent before execution. Quick starts differ by agent—/read for Claude, codex --full-auto with marketing-ops for Codex, auto-discovery on OpenClaw—but the pattern is the same: one entrypoint, many deep skills. Use it whenever you need credible marketing execution across phases, from validating messaging on a landing page to launch distribution, SEO architecture, lifecycle email, and ongoing CRO—not for writing application code or infra runbooks.2kinstalls2Engineering SkillsEngineering-skills is a meta package that catalogs twenty-three production-ready engineering skills for solo builders shipping with Claude Code, Codex, Cursor, Gemini CLI, OpenClaw, and related tools. Rather than one narrow task, it organizes senior architect, frontend, backend, fullstack, QA, DevOps, SecOps, code reviewer, and AI/ML or data engineering guides so you pick the right SKILL.md for the moment—from scaffolding a SaaS API to hardening CI/CD or reviewing a PR. Quick start documents /read paths and agent-skills-cli install for Codex. The bundle emphasizes stdlib-only Python utilities where tools are mentioned, which suits indies who want depth without sprawling dependencies. Treat it as a journey-wide engineering desk reference: invoke the umbrella when you are unsure which specialist skill fits, then drill into a child folder for concrete procedures.2kinstalls3Finance SkillsFinance Skills is a production-oriented financial analyst package for solo and indie builders who need spreadsheet-grade analysis inside an AI coding session. It centers on a financial-analyst skill folder plus four standard-library Python scripts so you can compute liquidity and profitability ratios, build discounted cash flow valuations, explain budget versus actual variance, and assemble rolling forecasts without wiring pandas or a BI stack. The meta SKILL instructs agents to pull only the specific sub-skill markdown needed for the task and to cross-check every number against your source statements, which keeps hallucinated metrics out of investor updates or pricing decisions. Use it when you are validating whether a SaaS price point covers costs, preparing a lightweight valuation narrative, or monitoring burn against plan during growth. It is a skill-and-script kit, not a live market data feed or tax advisor.2kinstalls4Business Growth Skillsbusiness-growth-skills is a MIT-licensed package of four production-oriented agent skills aimed at solo founders and small teams who still own customer success, deals, and rev ops without a full GTM org. Each sub-skill lives in its own folder with a dedicated SKILL.md, so you invoke customer-success-manager for health scoring and churn signals, sales-engineer for RFP and competitive matrices, revenue-operations for pipeline and forecast views, and contract-and-proposal-writer for proposals and contract templates. Nine companion Python scripts use only the standard library, which keeps installs lightweight on laptops and CI. Quick start paths include Claude Code reads and `npx agent-skills-cli` for Codex. The bundle fits when you are past prototype and need repeatable workflows for retention, enterprise-style sales motions, or board-ready pipeline numbers—not when you only need a single generic copywriting prompt.1.9kinstalls5Engineering Advanced SkillsEngineering Advanced Skills is a meta index and distribution bundle listing twenty-five POWERFUL-tier procedures for complex architecture, automation, and platform work—aimed at solo builders who outgrow single-purpose snippets and want named skills for agent design, API review, test generation, changelogs, pipelines, onboarding, schema design, and operational concerns. It does not replace reading each child SKILL.md; it routes you to the right document for Claude Code, Codex CLI, Gemini CLI, Cursor, or OpenClaw. Prism tags it under Build agent-tooling because installation and /read paths are the primary workflow, while many child skills naturally land in Ship (CI/CD, security) or Operate (monitoring, platform ops). Use this skill when you need a map of what exists in the engineering tree before committing to one implementation thread.1.8kinstalls6Product Skillsproduct-skills is a MIT-licensed product-team skill collection for solo and indie builders who need executive-style product discipline inside Claude Code, Codex, Cursor, Gemini CLI, or OpenClaw. Instead of one narrow integration, it packages roughly ten focused capabilities—prioritization with RICE, customer discovery and PRDs, agile backlog work, OKR and vision work, personas and journey maps, design tokens and component documentation, structured competitor analysis, landing page generation, SaaS project scaffolding, and research summarization—with optional Python utilities that stay stdlib-only. Use it when you are validating an idea or scoping a SaaS or content offer and want the agent to produce consistent PM and design artifacts rather than ad-hoc chat notes. It matters on Prism because it maps cleanly across Validate and Build phases while remaining a skill bundle rather than a hosted service, so buyers can judge fit by journey phase and deliverable type.1.8kinstalls7C Level Advisorc-level-advisor is a MIT-licensed virtual board ecosystem for founders and solo executives using Claude Code, Codex, Gemini CLI, Cursor, or OpenClaw. It packages many role-specific advisory skills behind a Chief of Staff that routes questions to CEO, CTO, finance, security, revenue, and people lenses, and it starts with /cs:setup to materialize company-context.md so every role shares stage, ARR or MRR, runway, and team facts. Big decisions can escalate to /cs:board for explicit multi-role debate with a built-in check that several perspectives responded before you treat a conclusion as final. Because strategic tradeoffs appear from first idea through growth and operations, the skill is journey-wide even though Prism shelves it under Idea research as the entry ritual. It is workflow-oriented advisory prose, not a deployment or MCP integration, so Prism users should match it to decision quality needs rather than codegen tasks.1.8kinstalls8Pm SkillsProject Management Skills is a six-pack of production-oriented agent capabilities for indie builders and tiny teams who still run formal delivery on Atlassian. Each sub-skill maps to a folder you load on demand: Senior PM for portfolio and risk framing, Scrum Master for sprint rhythm, Jira Expert for JQL and board automation, Confluence Expert for knowledge bases, Atlassian Admin for permissions and integrations, and Atlassian Templates for reusable blueprints. Claude Code users typically `/read` the specific SKILL.md; Codex users can install via agent-skills-cli. Optional MCP hooks let the agent query and update live Jira and Confluence instead of pasting stale exports. Companion Python utilities such as a project health dashboard script run without extra dependencies. Use the pack when you are the PM and the engineer and need agent-assisted planning, reporting, and Atlassian hygiene across build and operate.1.8kinstalls9Reviewreview is an agent skill for solo builders who rely on Claude Code auto-memory and need a structured audit instead of guessing what still belongs in MEMORY.md. It is shelved under Operate because the payoff appears once you have been iterating on a repo and memory files have accumulated noise, duplicates, and outdated tool references. The workflow locates the project memory directory, loads MEMORY.md, and scores entries for promotion candidacy, staleness, and merge opportunities; quick mode surfaces counts and top candidates when you only want a health snapshot. That matters for long-running agent projects where bad memory silently steers every session. The skill also respects the 200-line startup ceiling called out in the instructions, nudging you toward consolidation before context loading suffers. Use it after major refactors, dependency upgrades, or whenever the agent keeps repeating corrected facts.1.6kinstalls10Rememberremember is a journey-wide agent skill that writes deliberate knowledge into auto-memory when a discovery is too important to leave to implicit capture. Solo builders invoke it after debugging wins, undocumented conventions, tool-specific flags, architecture decisions, or personal coding preferences they want every future session to honor. The workflow parses the user’s statement into concrete facts and scope, searches existing memory for duplicates, then appends a timestamped entry under the project’s Claude memory directory. It fits anywhere in the solo journey—Validate when you lock scope assumptions, Build when you learn framework quirks, Ship when review reveals flaky test commands, Operate when production incidents surface root causes. Unlike stuffing CLAUDE.md for every micro-fact, remember targets durable session memory with minimal ceremony.1.6kinstalls11Ra Qm Skillsra-qm-skills bundles twelve regulatory affairs and quality-management agent capabilities aimed at solo founders and small HealthTech teams who must speak FDA, EU MDR, and ISO 13485 language without a full RA department. Each skill maps to a functional role—regulatory strategy, QMS governance, risk files, CAPA, controlled documentation, and audit readiness—backed by Python utilities that stay stdlib-only for simpler deployment in locked-down environments. ISO 27001 and GDPR skills extend the same pattern to information security and privacy for connected devices and SaaS wrappers. Use the bundle when shipment or EU/US market access depends on traceable QMS artifacts, not when you only need generic software linting. Installation differs by agent (slash-read in Claude Code, agent-skills-cli for Codex). Expect advanced domain literacy; the skills accelerate structured compliance work rather than replacing qualified human RA sign-off.1.6kinstalls12FixFix is an agent skill that walks solo builders through a flaky-test taxonomy for Playwright-style E2E suites. It starts from how a test fails—repeating locally, only in full suites, or only in CI—and maps each path to likely causes such as missing awaits, shared database state, viewport differences, or parallel worker limits. The readme embeds copy-paste before/after patterns so an agent can refactor tests toward deterministic assertions instead of arbitrary timeouts. Indie teams without a dedicated QA engineer can use it when CI goes red intermittently or when a green local run cannot be trusted. It complements init-style setup skills by stabilizing tests after infrastructure exists, and fits ship-phase quality work without replacing formal test design or coverage planning.1.6kinstalls13InitInit is an agent skill for setting up a production-oriented Playwright end-to-end stack in a project that may not yet have browser tests. It instructs the agent to scan package.json and tsconfig, avoid duplicating existing test dirs, install @playwright/test with browser deps, and emit a tailored playwright.config plus starter spec and CI hints. Solo builders shipping Next.js or SPA products use it when they say set up Playwright, add e2e tests, or configure test infrastructure—exact trigger phrases from the skill frontmatter. The outcome is repeatable local and CI runs with HTML reporting and sensible retries on CI rather than one-off manual installs. After init, flaky-test workflows or review skills help harden the suite; init does not replace writing meaningful user journeys or security testing.1.6kinstalls14Coveragecoverage is an agent skill that turns vague “are we tested enough?” anxiety into a concrete matrix for solo builders shipping SaaS, mobile shells, or extensions. When you say you want a coverage report or missing tests, the skill directs an Explore subagent to inventory every user-facing route, interactive component, API endpoint, and critical flow such as auth, checkout, or onboarding. It then parses your existing spec files, correlates page.goto calls, component locators, and mocked endpoints, and labels each area as covered, partial, or missing. The output is a markdown coverage matrix you can prioritize into new Playwright or similar tests during Ship testing sprints. It is multi-phase because you can run it mid-Build while features land, but the canonical shelf remains Ship testing. Complexity is intermediate: you need a test layout the agent can read and enough app structure to map. It does not execute tests or collect Istanbul percentages—it reasons about structural test presence, which is often what indie teams lack before launch.1.6kinstalls15GenerateGenerate is an agent skill packaged as concrete test-generation patterns for browser end-to-end coverage. It shows how to structure Playwright tests in TypeScript for high-traffic SaaS flows: signing in with valid and invalid credentials, asserting dashboard navigation and error alerts, and exercising create-and-edit paths on list UIs. Selectors lean on roles and labels rather than brittle CSS, which suits indie teams who want tests that survive modest markup tweaks. The readme is pattern-oriented rather than a full CLI scaffolder—you invoke it when you need copy-paste-ready describe blocks to extend your suite. It fits solo builders shipping web apps who already run Playwright but want consistent auth and CRUD templates from the agent instead of rewriting boilerplate each sprint. Pair it with your app's real routes and test data; extend patterns for delete, permissions, and API mocks as your ship checklist grows.1.6kinstalls16ReportReport is a smart test-reporting skill for solo builders who already use Playwright and want one agent-driven flow from execution to a shareable status update. When you say "test report", "results summary", or "how did tests go", the skill checks for existing `test-results/` or `playwright-report/` output, runs `npx playwright test` with JSON and HTML reporters if needed, then extracts passed, failed, skipped, flaky, and timing data from the JSON stream. It detects your destination automatically: TestRail via env configuration, Slack via webhook, GitHub Actions via workflow artifacts, or a served HTML report—and falls back to markdown when nothing else is wired. The design goal is zero new tools: reuse reporters and env vars you already have so shipping stays lightweight while stakeholders still get clarity on regressions.1.6kinstalls17ExtractExtract (/si:extract) is a journey-wide agent skill that turns a proven pattern or debugging solution into a standalone reusable skill with SKILL.md, reference docs, and examples. It is meant for moments when you have already fought through a non-obvious fix and want it portable across projects—after the same issue appears twice, when the solution is multi-step and easy to forget, or when you explicitly ask to save work as a skill. The workflow starts from your description, optionally mines Claude auto-memory via grep, then determines scope and scaffolds the skill package with dry-run preview support. Solo builders use it throughout the journey: after Ship debugging, during Build integrations, or in Operate when production quirks repeat. It is skill authoring meta-work, not runtime execution of the original fix.1.5kinstalls18PromotePromote is an agent skill for solo builders who use Claude Code long enough that good habits land in auto-memory but never stick in project rules. It walks through understanding the behavior you want enforced, locating matching lines in MEMORY.md under the project memory directory, confirming with you, then choosing the right sink—CLAUDE.md for repo-wide conventions or a scoped rules file when only certain paths matter. Targets can be specified explicitly (--target claude.md or rules/testing.md) and path globs can narrow scope. The outcome is durable instruction: background notes become first-class rules instead of hoping the model recalls them. It fits indie teams standardizing package managers, API validation, or test patterns after they have been proven in practice. Use it when a pattern has repeated successfully and you are ready to codify it; skip when you only want a one-off reminder without changing project config.1.5kinstalls19Statusstatus (/si:status) is a journey-wide agent skill that reports how healthy your Claude Code memory stack is before context rot slows every session. Solo builders who rely on MEMORY.md, topic files, project CLAUDE.md, and rules get line counts, topic sprawl signals, and a quick stale-path scan when entries point at deleted source files. The skill encodes explicit thresholds— for example MEMORY.md under 120 lines as healthy and CLAUDE.md under 150—so you can prune or split memory deliberately instead of guessing. It supports a one-line --brief mode for CI-style checks or a fuller dashboard when you are doing periodic maintenance. Because agent memory affects planning, implementation, shipping, and operations equally, the skill stays useful whenever a repository has accumulated months of agent context, not only on launch week.1.5kinstalls20Browserstackbrowserstack is an agent skill for solo builders who already use Playwright locally but must prove behavior on real Safari, Firefox, and varied OS targets without maintaining a device lab. It walks through verifying BrowserStack credentials, adapting playwright.config.ts with connectOptions and encoded capabilities JSON, and running the documented /pw:browserstack setup flow. The skill is intentionally gate-first: if username or access key env vars are unset, the agent tells you to fetch keys from BrowserStack account settings and pauses rather than failing obscurely mid-run. That fits indie teams shipping web SaaS or marketing sites who only pay for cloud minutes during Ship phase regression passes. It does not replace unit tests or security review; it extends your existing Playwright investment into a managed browser matrix when compatibility language appears in the task.1.5kinstalls21Migratemigrate is a Claude skill for solo builders and tiny teams stuck on Cypress or Selenium who want Playwright without rewriting every test from scratch in one chaotic session. It starts by detecting the source framework from directories, package.json, or Python imports, then produces a migration assessment with file counts and effort sizing so you pick sequential, batched, or parallel conversion. If Playwright is missing, it directs you through init first. Conversion is interactive and per file, loading dedicated mapping docs for Cypress commands, fixtures, and Selenium/WebDriver patterns. The workflow respects scope: small projects convert linearly, medium projects batch in fives with sub-agents, and large repos lean on batch parallelism. Triggers match natural language like migrate tests, switch to playwright, or move from cypress. Primary shelf is Ship → testing, but the same skill applies when you are modernizing E2E during Build refactors or pre-launch stabilization in Validate prototypes that already have legacy tests.1.5kinstalls22Testrailtestrail is an integration skill for builders who treat TestRail as the system of record while executing automation in Playwright. On import, it fetches cases via MCP, maps title, preconditions, steps, and expected results into Playwright tests grouped by section, and stamps each test with a TestRail annotation for traceability. On push, it runs Playwright with JSON output, parses pass and fail outcomes, and posts them to the designated run through TestRail's API via MCP. Triggers include mentions of test management, test cases, sync, and pushing results. Solo builders shipping web apps benefit when QA stakeholders live in TestRail but engineering lives in the repo. Canonical Prism placement is Ship → testing because the highest-value moment is closing the loop on a run; import supports Build when you are expanding automated coverage from manual case libraries. The skill expects credentials in environment variables and will not proceed silently without them.1.5kinstalls23LoopLoop is a Claude slash-command skill that wires an autonomous experiment loop to a user-chosen cadence. Solo builders who track hypotheses as experiment paths—API speed, reliability gates, agent evals—can start background iteration instead of re-running the same prompt manually. You invoke /ar:loop with an experiment identifier and optional shorthand (10m, 1h, daily, weekly, monthly) or use interactive picks when arguments are missing; stopping is explicit with /ar:loop stop. The skill resolves the experiment, translates the interval to a cron schedule, and relies on CronCreate so the agent revisits the experiment on that rhythm. It fits indie teams who already organize work under experiment folders and want agent time on a clock, not only on demand. It is automation for sustained measurement and fix attempts, not a one-shot code generator.1.4kinstalls24ResumeResume is an agent skill for solo builders running autoresearch-style experiments who hit context limits or intentionally paused a branch. Invoked with /ar:resume, it either lists experiments for you to pick or jumps straight to engineering/api-speed-style paths. The agent runs the setup script, checks out the autoresearch branch, reads config and program strategy, replays results.tsv history, and summarizes keeps, discards, crashes, and the current best metric before continuing iteration. It assumes a git-backed experiment layout under .autoresearch and pairs naturally with metric-driven change proposals on a single target file. Best for indie hackers optimizing APIs or hot paths who do not want to re-derive state from chat memory. Not a substitute for initial experiment setup or greenfield implementation planning.1.4kinstalls25SetupSetup is an agent skill that registers a new autoresearch experiment so Claude or similar agents can iterate on a file against a measurable eval. You can pass domain, experiment name, target path, eval command, metric name, and whether lower or higher is better in one invocation, or walk through each field interactively when you run /ar:setup with no args. It supports listing existing experiments and built-in evaluators, optional evaluator selection, and scope hints for how broadly the optimizer may change code. Solo builders use it when standing up API-speed benches, content CTR tweaks, or prompt variants without hand-writing YAML for every trial. The skill is the front door to the autoresearch workflow: once configured, downstream optimize runs have a single source of truth for what to edit and how to score outcomes.1.4kinstalls26RunRun is an agent skill that executes exactly one autoresearch experiment iteration for solo builders who treat improvement as a logged science project rather than random tweaks. It wires into the /ar:run command, resolves which experiment to use, loads constraints and history from .autoresearch/{domain}/{name}, checks out the dedicated git branch, and forces a single deliberate edit to the configured target file before evaluation. The workflow explicitly reviews what was kept, discarded, or crashed in results.tsv and escalates strategy from low-hanging fruit through structural and radical attempts as run count grows. That structure suits indie operators optimizing APIs, pipelines, or other measurable targets without losing reproducibility. It is not a bulk code generator; it is a tight iteration ritual meant to compound learning across dozens of runs while respecting program.md guardrails.1.4kinstalls27Aws Solution ArchitectAWS Solution Architect is an agent skill that emits structured architecture recommendations for solo builders planning a serious SaaS or API on AWS. Instead of improvising service names in chat, you receive a named pattern, mapped components, cost bands, scaling characteristics, and a week-by-week implementation plan spanning VPC foundation through blue/green CI/CD. It sits early in the build phase when you are choosing backend shape, but the same output informs validate-time scope checks and operate-time infra expectations. Prism catalogs it for indie teams who need a defensible default before they touch the console or Terraform. Treat dollar figures as starting estimates—region, redundancy, and data size will move the bill. Use the phased task lists as checkpoints against your own compliance and observability requirements.1.4kinstalls28PostmortemPostmortem is a journey-wide agent skill that guides honest, system-level analysis after something goes wrong. It targets solo founders and indie teams who would otherwise either scapegoat someone or publish twelve fuzzy “do better” bullets. The framework starts by defining the event precisely—expected outcome, actual outcome, when the gap appeared, and financial or reputational impact—then investigates conditions that made failure predictable in hindsight, not who to blame. Invoke it after a failed deal, missed quarter, feature that did not land, or hire that did not work out, as well as after production incidents or launch disappointments. The output should be actionable system changes, not performative retros. It complements planning skills you run before work; this one closes the loop when reality diverged from the plan.1.4kinstalls29Board PrepBoard Prep is an agent skill for solo founders and small-team CEOs facing board meetings, investor updates, fundraising presentations, or any high-stakes adversarial review where credibility depends on knowing your numbers and telling the truth about risk. It implements `/em:board-prep` with a preparation framework centered on Numbers Cold—every figure in the deck should be memorized, not read—and on anticipating the questions experienced directors ask after seeing dozens of companies. The skill pushes a narrative that surfaces weakness clearly while keeping confidence in your plan and execution, avoiding sanitized decks that board members instantly distrust. Canonical placement is validate pricing because runway, growth, churn, and fundraising story are under the microscope, though operating CEOs reuse it before quarterly board cycles. It is editorial and strategic workflow content, not slide design software; pair it with your existing deck and data exports. Use when the SKILL description calls for board meeting preparation where flinching at your own metrics costs trust.1.4kinstalls30SpawnSpawn is a hub command skill that launches multiple subagents to work the same session task in parallel, each isolated in its own git worktree so branches do not stomp each other. Invocation supports bare `/hub:spawn`, a specific session timestamp, and `--template` to pull structured dispatch prompts for optimizer, refactorer, test-writer, or bug-fixer modes—each with a repeatable loop (eval, tests, coverage, or reproduce-fix-verify). The workflow loads session config, substitutes template variables, and encourages different competitive strategies per agent to maximize exploration value. Solo builders use it when one agent path might be slow or wrong: performance tuning, refactors, test gaps, or ambiguous bugs. It sits in build-phase agent-tooling and extends naturally into ship-phase testing and review when agents race toward green CI. It assumes a hub-style session layout and git; it does not replace code review policy or production deploy gates.1.4kinstalls31Boardboard exposes the AgentHub message board so solo builders running multi-agent evals or parallel tasks can assign work, surface status, and collect merge-ready results in one filesystem-backed convention. Channels separate concerns: dispatch carries coordinator assignments, progress holds agent updates, and results stores finals plus merge summaries. Operators list and read channels in order, post new messages with author attribution, or reply in threads tied to post IDs. It is intentionally lightweight markdown—not a realtime chat server—making it easy to grep, version, and audit alongside your repo. Use it when one Claude session plays coordinator and others execute, and you need durable handoffs that survive context limits better than scrolling chat history.1.4kinstalls32Stress Teststress-test is a journey-wide agent skill from the /em command family that helps solo founders and indie builders falsify business assumptions while they are still cheap to change. Instead of debating whether "the market is big," you state a precise claim—such as a €2.3B SAM for German SME spend management—and run a structured break-it-first review across market size, willingness to pay, churn, expansion, hiring velocity, and revenue timelines. The methodology treats optimism as necessary to start but dangerous when embedded unchecked in spreadsheets everyone nods along with. Use it whenever a plan, pitch deck, or financial model rests on consensus numbers, before you commit roadmap, hire ahead of revenue, or enter a fundraise process. Outputs are sharper, testable hypotheses and identified failure modes—not a doom narrative. It pairs naturally with ideation and validation work and remains useful in Grow when revisiting retention or expansion bets.1.4kinstalls33ChallengeChallenge is a journey-wide agent skill for solo and indie builders who need disciplined skepticism before locking in a plan. Invoked as /em:challenge on a plan document or pitch, it applies the pre-mortem method: assume the initiative failed a year from now and enumerate why—overestimated demand, hidden complexity, timing, and dependency risks. The SKILL.md positions it before major resource commits, investor or board presentations, fast-move pressure, or stretches of only positive feedback. It pairs naturally with ideation and planning skills but stands alone whenever a decision deserves a structured adversarial pass. Output is a clearer mental model of failure modes so you adjust scope, sequencing, or validation steps instead of discovering the same gaps in production.1.4kinstalls34Hard Callhard-call is an agent skill packaged as the `/em:hard-call` command: a decision framework for moments when every option hurts—terminations, layoffs, product kills, pivots, and shutdowns. It assumes the data is often already clear and that difficulty comes from human impact, avoidance, irreversibility, and your own emotional stake. The workflow starts with a reversibility test, then walks through structured steps to surface a less-wrong choice instead of chasing a fictional right answer. Solo and indie builders wear every hat, so these calls land without an executive coach in the room; the skill gives your coding agent repeatable language and checkpoints you can run at 3am before you send the message or flip the strategy. Use it journey-wide whenever you have been postponing a call you already know you need to make, not only during a crisis week in production ops.1.4kinstalls35Evaleval is an AgentHub-oriented agent skill invoked as /hub:eval to compare every agent result in a session and declare a winner. Solo builders running multi-agent experiments on one task use it when subjective chat opinions are not enough—you need repeatable ranking either from an executable metric (latency, test pass rate, custom script output) or from an LLM judge that reads git diffs and agent write-ups. Metric mode shells into each worktree with result_ranker.py; judge mode emphasizes correctness first, then fewer changed lines when correctness ties, then execution quality. It assumes an existing AgentHub session with branches and result artifacts, not greenfield coding from scratch. Use it after agents finish and before you merge or promote one branch, especially when you are A/B testing prompts, skills, or tool configurations.1.4kinstalls36MergeMerge is an agent skill for Agent Hub multi-agent coding sessions: after agents compete on isolated branches, you invoke `/hub:merge` to land the winning attempt on your base branch, preserve losing work as archive tags, and remove ephemeral worktrees. Solo builders running several Claude or Codex agents in parallel use it to avoid manual git gymnastics—checkout base, merge the ranked winner from `/hub:eval`, tag `hub/archive/...` for every non-winner, delete loser branch names, and run the bundled session manager cleanup script. You can target the latest session, a timestamped session id, or force a specific agent id when human judgment disagrees with the leaderboard. The flow assumes hub-style branch naming (`hub/{session-id}/{agent-id}/attempt-1`) and produces a markdown merge summary on the results board. It is operational git workflow knowledge, not application feature code—best paired with eval ranking before merge and clear task descriptions in merge commit messages for auditability.1.4kinstalls37Social Media AnalyzerSocial Media Analyzer is an agent skill that structures social campaign results into a review-ready JSON report: platform-level metrics, ROI economics, top-performing posts, and benchmark-backed insights. Solo and indie builders who run paid or organic social without a dedicated growth analyst can invoke it after a campaign window closes to see whether engagement and CTR beat typical benchmarks and whether spend justified estimated value. The skill emphasizes overall health labels, strengths (engagement, ROI, CTR), and concrete next steps such as increasing budget or cloning successful post patterns to other campaigns. It suits founders shipping content-led SaaS, newsletters, or indie products who need repeatable post-mortems instead of spreadsheet wrangling. Pair it with your own exported metrics or API pulls from the platform; the skill focuses on interpretation and recommendation logic rather than live OAuth wiring.1.2kinstalls38Senior FrontendSenior-frontend encodes a decision profile for solo and small teams building content-first properties—marketing pages, documentation, blogs, and pricing sites where reads dominate and every JavaScript byte must earn its place. The skill materializes as a structured astro-or-static schema: ranked framework options (Astro, Eleventy, Hugo, Next static export), TypeScript or markdown-first authoring, Tailwind or CSS modules, MDX in-repo or headless CMS, and islands-only client hydration. It flags hard kills such as SPA architectures for SEO-critical marketing and blanket React hydration, plus warnings on Google Tag Manager versus Core Web Vitals. Testing expectations include Playwright end-to-end runs and Lighthouse CI. Invoke it when an agent is about to scaffold or refactor a public site and you want enforceable constraints instead of generic React defaults. It targets builders shipping on mobile 4G and desktop fiber who depend on static generation plus selective hydration for 2026 search and AI-search visibility.1kinstalls39Senior ArchitectSenior-architect is a reference agent skill that walks solo and indie builders through major software architecture patterns with the problems each solves, fit criteria, and honest trade-offs. It is aimed at people who must choose a shape for a SaaS, API, or agent-backed product without a staff architect on call. Use it when you are scoping an MVP, refactoring toward clearer boundaries, or debating whether event-driven or CQRS is worth the ops tax. The guide indexes nine patterns—including monolithic, modular monolith, microservices, event-driven, CQRS, event sourcing, hexagonal, clean architecture, and API gateway—and pairs each with team-size guidance, scaling implications, and structural sketches. It does not replace capacity planning or cloud cost modeling, but it gives your agent consistent vocabulary for proposals, ADRs, and implementation plans. Intermediate-to-advanced readers get the most value; beginners should start with monolith and modular monolith sections before distributed options.996installs40Research SummarizerResearch Summarizer equips coding agents to produce research summaries backed by disciplined citation formatting, centered on APA 7 for business and social-science style sources and IEEE for technical papers. Solo builders validating markets, writing positioning docs, or drafting investor appendices use it when raw chat paraphrases are not enough and every claim needs a traceable reference line. The bundled quick reference walks through journal, book, conference, and web page patterns plus in-text citation variants so outputs read like professional briefs rather than unattributed bullet lists. It pairs naturally with early journey work—competitor teardowns, audience interviews distilled from published studies, and technical due diligence—while remaining useful in Validate when you tighten scope with cited evidence. Treat it as the citation layer on top of your search workflow, not a substitute for verifying primary sources yourself.967installs41Senior BackendSenior-backend is an agent skill that encodes a batteries-included Django monolith blueprint as a structured profile agents can follow when scaffolding or reviewing server-side work. It is aimed at solo builders and small teams who would otherwise hand-roll auth, admin, APIs, and background jobs inconsistently across a content-heavy or marketplace-style SaaS. The profile pins framework, language, database, cache, queue, auth, testing (pytest, factory-boy), and when HTML is needed (templates or htmx). Constraints spell team size, tenancy, data sensitivity, and that an admin panel is expected. Anti-recommendations act as hard editorial gates—one framework stack, no thread hacks for async work, rate limits required, monolith over premature microservices. Use it during backend design and implementation reviews so agent output stays aligned with a proven CRUD-first architecture rather than trendy splits that do not match the scale.964installs42Tdd GuideTDD Guide is an agent skill for solo builders who want test-driven development to feel mechanical rather than aspirational. It helps you spin up focused unit tests—happy paths, error cases, and priority-tagged cases—then interpret coverage and complexity so you know what still risks breaking in production. The documented flow mirrors classic red-green-refactor: define failing expectations, implement the smallest pass, then refactor with metrics still green. It is especially useful when you are moving fast on auth, validators, or service layers where branch coverage hides bugs. Primary placement is Ship and testing, but you will invoke it throughout Build as you add modules. Pair it with your normal CI habit; the skill accelerates authoring and gap spotting, not replacing your judgment on what deserves an integration test.907installs43Code ReviewerCode Reviewer is an agent skill for solo builders maintaining C codebases who want reviews grounded in shared rule files rather than vague "looks fine" feedback. It applies universal review rules together with C-specific guidance to catch classic issues: unbounded copies, missing NULL checks after malloc, dangerous format-string usage, and command-injection surfaces via system(). The bundled narrative walks a deliberately smelly sample through a clean refactor so you see what compliant patterns look like in practice. Use it during active implementation when you want a second pass on safety, or during Ship when you are hardening a CLI, embedded helper, or native extension before release. It is narrower than a full static-analysis pipeline but faster to invoke inside Claude Code or Cursor when you already organize standards as markdown rules.879installs44Senior FullstackSenior Fullstack is an agent skill that combines fullstack project scaffolding with code quality analysis for indie builders shipping web apps. It responds when you ask to scaffold a new project, create a Next.js app, wire FastAPI with React, generate MERN or Django-React boilerplate, analyze code quality, audit a codebase, or compare tech stacks. Supported scaffolds include Next.js with App Router, FastAPI paired with React and PostgreSQL, MERN with TypeScript, and Django REST with React. Workflows cover listing templates, generating structures, and running quality checks framed around security and complexity. Use it early in build to avoid weeks of ad-hoc folder sprawl, during validate when stack choice is still open, and before ship when you want a structured pass over security smells. The skill is intermediate because you still own deployment, env secrets, and production hardening after boilerplate lands.872installs45Senior QaSenior QA is a production-oriented agent skill for solo and indie builders shipping React and Next.js products. It bundles procedural knowledge plus runnable Python helpers that scan component trees, emit Jest and React Testing Library test stubs (including accessibility assertions), scaffold Playwright end-to-end suites with page objects, and analyze Istanbul or LCOV reports against a threshold you set. The workflow fits when you have UI code but thin or missing tests, when you need a repeatable coverage gate before merge or deploy, or when you want MSW-aligned mocking and a11y checks without assembling the toolchain from scratch. Use it during Ship for testing discipline; it also supports Build when you are hardening a feature branch or documenting QA expectations for a small team of one.853installs46Ui Design Systemui-design-system is an agent skill that supplies a fill-in design-system documentation template for solo and indie builders who need a single source of truth for visual language. It walks through system metadata, four guiding principles, and tabular brand and neutral palettes so Claude Code, Cursor, or similar agents can extend the same structure into typography, spacing, and components without inventing ad hoc naming. Use it when you are starting or refreshing a product UI and want principles—especially accessibility-by-default—to precede one-off screen tweaks. It matters because inconsistent tokens and undocumented grays are what make AI-generated interfaces look generic; a written system gives agents concrete constraints. Pair it with your actual component library or Figma tokens once the tables are populated.853installs47Senior Computer VisionSenior Computer Vision is an agent skill that walks solo builders through production-grade visual AI: choosing detectors and segmenters, preparing labeled data, training with mainstream frameworks, and squeezing latency for real deployments. It targets founders and indie engineers who need a credible CV stack without hiring a full ML team—whether that is a SaaS feature, an internal automation agent, or an API that scores images. Invoke it when you are building detection pipelines, training custom weights, optimizing inference, or packaging models for edge or cloud. The skill pairs conceptual architecture guidance with script-oriented quick starts so your coding agent can generate YOLO or Faster R-CNN configs, run optimization benchmarks toward ONNX, and structure dataset augmentation pipelines. Expect intermediate complexity: you should be comfortable with Python and basic ML concepts, but the skill reduces thrashing across Ultralytics, Detectron2, and deployment tooling.843installs48Senior Data Engineersenior-data-engineer is a production-oriented agent skill for scalable pipelines, ETL/ELT, and data infrastructure. Solo builders who outgrow ad-hoc SQL exports invoke it when designing architectures, building ingestion, optimizing workflows, implementing governance, or debugging pipeline failures. The SKILL.md organizes trigger phrases for pipeline and architecture questions, quick start paths, workflows, an architecture decision framework, tech stack notes, reference docs, and troubleshooting. Expertise spans Python, SQL, Spark, Airflow, dbt, and Kafka, with guidance on star and snowflake schemas, slowly changing dimensions, data vaults, batch versus streaming, lakehouses, and data quality checks. It is multi-phase: most visible during build when backend data layers are created, but equally relevant in operate when pipelines break and in grow when analytics and lifecycle metrics depend on trustworthy tables. Intermediate to advanced complexity reflects real orchestration and distributed processing choices rather than one-off scripts.842installs49Senior Prompt EngineerSenior Prompt Engineer is an agent skill for solo and indie builders who need repeatable patterns instead of one-off chat experiments. It bundles prompt optimization, RAG evaluation, and agent-orchestration utilities with workflow guides for few-shot examples, structured outputs, and token-aware design. Use it when you are wiring Claude Code, Cursor, or Codex into features that must return reliable JSON, pass retrieval benchmarks, or chain multiple model steps. The skill fits early agent design during Build, but the same checklists help you re-tune prompts before launch copy, validate prototype demos, and regression-test model behavior before Ship. Expect intermediate depth: you run Python scripts locally, maintain prompt and context files, and interpret evaluation metrics yourself. It is a skill package with reference docs, not a hosted MCP server or managed eval platform.819installs50Senior Securitysenior-security packages a senior security engineer playbook for solo and indie builders who must threat-model, review code, and respond to incidents without a dedicated AppSec team. The skill organizes repeatable workflows: STRIDE threat modeling from scope definition through mitigation planning, architecture hardening including zero-trust framing, structured vulnerability assessment, secure code review checklists, and incident response steps. Triggers span architecture reviews, cryptography implementation questions, attack surface analysis, and audit requests, so the agent does not treat security as a one-off lint pass. Use it when shipping customer data, exposing APIs, or integrating third-party auth—and again in Operate when investigating suspected breaches or prioritizing CVE fixes. Outputs are analysis artifacts and prioritized remediation guidance rather than automated scan dumps; pair with repo-specific scanners where the skill references external tools.815installs51Ceo AdvisorCEO Advisor packages board governance and investor relations guidance for founders who suddenly need institutional process without a corporate secretary on staff. The skill reads like an executive playbook: how to size and diversify a board, what each leadership role owns, and how to run a four-meeting annual calendar spanning strategy, results, mid-year planning, and budget or comp approval. Solo builders nearing seed-plus financings or formal advisors can use it to draft agendas, pre-read timelines starting four weeks out, and committee charters for audit, compensation, and nominations. It does not replace lawyers on securities compliance; it gives language and structure for meetings, evaluations, and succession conversations. Pair it with your cap-table reality—many sections assume a scaling company rather than a weekend side project.807installs52Senior DevopsSenior DevOps is an agent skill packaged as a CI/CD pipeline reference guide for solo and indie builders who want consistent DevOps guidance instead of one-off chat answers. It walks through repeatable patterns—from baseline best-practice implementations through more advanced async-style examples—plus guidelines on how to structure automation code, reason about bottlenecks, and keep security in scope while pipelines evolve. The skill is meant to be invoked when you are designing or refactoring delivery workflows, hardening release paths, or teaching your agent how senior engineers typically trade off benefits and operational costs. Because shipping and operating a product both depend on the same automation choices, treat this as a multi-phase shelf item: primary home in Build integrations, with natural follow-on use during Ship and Operate when you tune performance and monitoring. Pair it with concrete repo context (existing CI files, cloud target) so the patterns map to your stack rather than staying abstract.806installs53Product Manager ToolkitProduct Manager Toolkit gives solo builders a structured PRD template an agent can fill in collaboratively: document metadata, a crisp problem statement backed by interviews or analytics, prioritized user stories, solution overview with flows, and success criteria. It forces you to separate pain from solution, name who suffers, cite evidence, and quantify inaction—habits that prevent scope creep when one person plays PM and engineer. The artifact is meant to move from Draft through Approved with named reviewers and a target release, so you can hand the same doc to an implementation agent or a contractor. It sits early in the journey when you are proving what to build, then remains the north star through Build and Ship. Intermediate complexity because you need real customer or data signals—not placeholder lorem—to get value from the evidence and success sections.800installs54Agile Product OwnerAgile Product Owner is an agent skill built around a sprint planning document template for solo builders and tiny teams who still want Scrum-style rigor without enterprise overhead. It walks you through defining a one-sentence sprint goal, documenting team capacity against historical velocity, selecting stories with story points and priorities, and surfacing cross-team dependencies before work starts. The output is copy-paste-ready markdown you can drop into Notion, GitHub, or Linear as the source of truth for a sprint. It fits when you are committing to a two-week (or similar) delivery window and need everyone aligned on what must ship versus nice-to-have scope. It does not replace discovery or GTM planning—it assumes you already have stories or tickets to pull from the backlog.798installs55Ux Researcher DesignerUX Researcher & Designer is an agent skill that packages senior UX researcher workflows for solo and indie builders who need structured research artifacts without hiring a full research team. Invoke it when you have raw research—interviews, surveys, analytics snippets—and need personas, journey maps, usability test plans, or synthesized findings. The skill walks through four explicit workflows: persona generation from data, journey mapping, usability test design, and cross-study synthesis. It aligns with Prism’s journey from early audience understanding through validation, and it supports design decisions later in Build when you need to ground UI choices in stated user needs. Use it whenever trigger phrases like “create user persona,” “map user journey,” or “synthesize interview findings” match your task. It is procedural knowledge for agents (Claude Code, Cursor, Codex), not a live research platform or recruiting service.795installs56Karpathy CoderKarpathy Coder is an agent skill that turns Andrej Karpathy’s LLM coding observations into enforceable practice for solo builders shipping with Claude Code, Cursor, or Codex. It addresses the common failure mode where models assume context silently, overbuild abstractions, leave dead code, and loop without measurable success criteria. The package pairs procedural rules with runnable Python checks, a dedicated review agent, a slash command, and a pre-commit hook so discipline shows up at review and commit time—not only in chat. Use it when you are writing or refactoring, when a diff feels too large for the task, when you suspect unnecessary layers or unclear goals, or when you want a structured “push back” pass before you ship. It fits SaaS backends, CLIs, and API services where small teams cannot afford review debt. Version 2.3.0 ships under MIT with tags spanning code-quality, simplicity, and surgical-change anti-patterns.794installs57Tech Stack EvaluatorTech Stack Evaluator is an agent skill that turns fuzzy “Postgres or Mongo?” debates into a structured scoring model solo builders can cite in ADRs and pitch decks. For each candidate technology it produces category scores across eight facets, rolls them into a weighted total, surfaces strengths and weaknesses, and explains which dimension drove the recommendation with explicit importance weights. The sample output illustrates PostgreSQL versus MongoDB with differentiated winners on scalability and developer experience—useful when you are validating an MVP data layer, planning a SaaS backend, or revisiting stack debt before a growth push. It is not a substitute for load tests or vendor quotes; treat confidence percentages as model output and validate against your traffic, team skills, and compliance needs. Pair it with prototyping skills once a shortlist is clear.792installs58Content CreatorContent Creator is a deprecated redirect skill for agents still using the old “content creator” name. Instead of writing or planning in one bundle, it reads user intent and points to content-production for blog posts, articles, guides, brand voice analysis, and on-page SEO tooling, or to content-strategy for topic clusters, calendars, and editorial planning. Social content requests go to social-content. Solo builders benefit because the agent stops blending strategy and execution in a single vague skill. Install it only if your workflow or docs still reference content-creator; otherwise invoke the specialist skills directly for faster, more accurate outcomes.786installs59Self Improving AgentSelf-Improving Agent is a journey-wide Claude Code companion skill that adds curation on top of auto-memory. While v2.1.32+ auto-memory continuously appends project patterns and preferences to MEMORY.md, this skill supplies structured commands to review what accumulated, promote durable rules into CLAUDE.md or scoped .claude/rules/, and extract repeated fixes into installable skills. Solo builders benefit because agent sessions otherwise re-learn the same conventions every week. Use /si:review when MEMORY.md feels noisy, /si:promote when a note has survived three real tasks, /si:extract when a debugging ritual should become a skill, and /si:status before large refactors to avoid blowing context on redundant notes. It fits any phase where you rely on Claude Code in the same repository—idea research scripts, validate spikes, build features, ship reviews—because memory hygiene is orthogonal to product stage. The skill is procedural and meta: it does not replace brainstorming or planning skills, but makes their outputs stick as enforced project knowledge.782installs60Senior Data ScientistSenior Data Scientist packages opinionated frameworks for experiment design, feature engineering, and operating ML in production. Solo builders shipping analytics-heavy SaaS or API products can invoke it when moving from a hypothesis to engineered features and deployable models without reinventing senior-team playbooks. The skill stresses production-first constraints—throughput, fault tolerance, encryption, access control, and audit logging—alongside code quality habits like type hints, profiling before optimization, and strategic caching. It spans conceptual design for experiments and the implementation patterns for batch and real-time pipelines, making it useful whenever an agent drafts architecture, reviews a notebook for production gaps, or documents an ML service boundary. It is advanced material: expect dense patterns rather than a single copy-paste script.774installs61Senior Ml EngineerSenior-ml-engineer in this bundle presents an LLM Integration Guide for solo builders shipping AI features—not generic ML training coursework. It walks through production-facing concerns: a provider abstraction layer so you can swap OpenAI and Anthropic implementations, prompt engineering discipline, token optimization, cost management, and error handling when APIs fail or rate-limit. The material suits backend or full-stack indie products adding chat, completion, or agentic flows behind a stable interface. Use during build when connecting models to your app, and revisit during operate when tuning spend and reliability. Intermediate to advanced depending on whether you already run Python services and manage API keys securely. Pair with security and secrets practices before exposing keys in agent sessions.770installs62Senior SecopsSenior SecOps is a compliance reference skill for solo and indie builders who sell SaaS or APIs into regulated or security-conscious markets. It packages structured guidance for SOC 2 Type II trust criteria, PCI-DSS, HIPAA, and GDPR, including security control checklists for control environment, communication, risk assessment, and related domains. Use it when you need to document policies, assign responsibilities, and prepare evidence trails before an audit or enterprise security questionnaire—not as a substitute for legal counsel or a formal GRC platform. The skill emphasizes practical markdown checklists and automation-oriented audit prep so an agent can help you gap-assess controls, align monitoring and encryption narratives, and iterate as the product ships and operates. It fits builders wearing both engineering and security hats who must move from ad-hoc notes to auditor-friendly structure without a dedicated SecOps team.768installs63Gdpr Dsgvo ExpertGDPR/DSGVO Expert packages compliance automation for solo builders shipping SaaS, APIs, or ecommerce to EU users. It centers on a Python GDPR Compliance Checker that walks project trees for emails, phones, IPs, health or biometric hints, payment data, and anti-patterns like logging PII, missing consent, indefinite retention, weak encryption, or broken erasure paths—then emits a scored report suitable for local review or CI JSON artifacts. Companion flows cover DPIA documentation and tracking data subject access or deletion requests, aligned with GDPR and German BDSG expectations. Use it before launch, after major data-model changes, or when an enterprise prospect asks for privacy evidence. It augments legal judgment; it does not replace counsel—but it gives indie teams a repeatable technical baseline agents can run in Ship and Operate.766installs64Product Strategistproduct-strategist packages an OKR planning template for solo builders and small product leads who need disciplined scope without a full PMO. The skill surfaces a fill-in structure: planning metadata, a company-level objective, then one or more product objectives each with three key results, baselines, targets, progress columns, and linked initiatives. That format helps you validate whether a quarter’s bets are measurable, owned, and sized before you write implementation plans or ship features. It is best invoked when you are choosing what to build next, aligning a micro-SaaS roadmap to outcomes, or preparing investor or team check-ins. The agent uses the template to draft or refine objectives and KRs in plain tables rather than ad-hoc bullet lists, keeping strategy artifacts consistent enough to revisit weekly or monthly without re-deriving structure from scratch.759installs65Risk Management Specialistrisk-management-specialist is an agent skill that walks you through implementing medical device risk management per ISO 14971:2019. It is aimed at solo builders and small teams shipping regulated health software who need audit-ready structure without a full regulatory department. Use it when you must define scope and acceptability criteria, run risk analysis and evaluation, document controls and residual risk, and maintain post-production surveillance. The guide organizes planning through reporting with explicit documentation expectations and a reusable risk management plan template. It matters because scattered notes fail notified-body and QMS reviews; a consistent ISO-shaped artifact chain reduces rework before Ship security work and ongoing Operate iteration.757installs66Social Media ManagerSocial Media Manager is an agent skill backed by social_calendar_generator.py, a 100% standard-library script that turns a small JSON config into a multi-week editorial calendar for solo and indie builders who need consistent posts without a SaaS scheduler. You define content pillars with emoji, descriptions, and weights, then assign platforms with weekly post counts and preferred weekdays; the generator balances pillar mix across the horizon and can run in demo mode when you are exploring formats. Outputs suit Claude Code or Cursor workflows: pipe --markdown into docs, feed --json to downstream copy skills, or iterate pillars before you commit to a launch or growth sprint. It does not write post copy or attach analytics—it plans when and what theme to publish so you can pair it with drafting and distribution tools in the Grow phase.755installs67Cto AdvisorCTO Advisor is an agent skill that equips solo and indie builders with a rigorous Architecture Decision Record (ADR) framework instead of leaving important choices buried in chat logs. It defines what an ADR is, supplies a complete markdown template with status, deciders, context, decision drivers, considered options, chosen outcome, pros and cons per option, and cross-links to related records and research. That structure is ideal when you are picking monolith vs services, auth strategy, hosting, or data stores and need a durable artifact you can hand to a future collaborator—or your future self. Use it whenever a decision has long-lived consequences, not for trivial naming bikesheds. The skill is prose- and template-driven rather than a live architecture scanner; you still own verification and implementation. Pair ADRs with your repo docs or a /docs/adr folder as your product grows.754installs68Marketing Strategy Pmmmarketing-strategy-pmm packages international go-to-market playbooks for solo founders and small PMM teams stretching into US, UK, DACH, France, and Canada without a full localization agency on retainer. It sits at launch when distribution strategy must turn into sequenced market entry: which country gets budget first, what ARR target fits each phase, and what must be true before you enter. The readme emphasizes a Series A expansion table, market readiness gates including three inbound leads in ninety days, and country-specific buying and communication norms starting with the US. Use it when you have an initial product and need structured PMM guidance rather than ad-hoc positioning chats. It also supports grow-phase planning when you rebalance channels or add regions after first revenue.749installs69App Store OptimizationApp Store Optimization is an agent skill packaged as a systematic ASO audit template for solo and indie builders shipping iOS or Android apps. It walks you through capturing app context, then scoring listing metadata against platform character limits, keyword placement, brand positioning, and benefit-focused copy. The workflow separates iOS and Android criteria so you do not mix subtitle rules with short-description rules or waste the iOS keyword field on duplicates and plurals. Each section ends with a numeric score and empty recommendation lines your agent can fill after competitor and keyword research. Use it when you are preparing a new store listing, refreshing copy after a rating dip, or validating scope before a launch sprint. It complements human keyword tools and analytics; it does not replace live store experiments or Apple or Google policy review. Output is an editorial-ready audit document your team can turn into listing updates and A/B hypotheses.744installs70Marketing Demand AcquisitionMarketing Demand & Acquisition is a structured marketing skill for solo operators and small growth teams at Series A and beyond who must plan demand generation without a full agency bench. It walks through core KPIs, a demand-gen framework, channel-specific paid media guidance, SEO, partnerships, attribution, and tooling references—oriented to hybrid product-led and sales-led motions in EU, US, and Canada. Invoke it when you are setting acquisition strategy, sizing budgets, optimizing PPC, designing lead and pipeline workflows, or aligning HubSpot-style campaigns with CAC and MQL/SQL definitions. The content is strategic and playbook-driven rather than a single ad connector, so it fits planning sessions before you spend or before you rewrite landing and SEO surfaces for growth.737installs71Quality Documentation ManagerQuality Documentation Manager is an agent skill for founders and quality leads building or maintaining a medical device quality management system where every controlled document must be traceable. It walks through document control from assignment of numbers and template use through review, approval, change control, and retirement, aligned with ISO 13485 expectations and U.S. electronic record rules under 21 CFR Part 11. Triggers such as document numbering, version control, change control, electronic signature, and audit trail map directly to operational pain: teams otherwise paste generic wiki advice that fails auditor questions on master lists and retention. Use it when you are formalizing SOPs, design history files, or software lifecycle records—not when you only need a public API README. The skill is procedural and checklist-oriented across multiple TOC sections rather than a single automation script, so agents produce coherent QMS language and workflows you still validate with regulatory counsel. It complements security and review skills by focusing on documentation lifecycle integrity instead of penetration testing or code defects.736installs72Regulatory Affairs HeadRegulatory Affairs Head is an agent skill packaged as an EU MDR 2017/745 submission guide for teams building regulated medical devices, including software that falls under MDR scope. It organizes the conformity landscape from Class I self-certification under Annex II through Class IIa, IIb, and Class III notified-body pathways involving type examination, full quality assurance, and production surveillance. The content emphasizes what must appear in technical documentation—device description, intended purpose, ISO 14971-aligned risk management, clinical evidence per Annex XIV, post-market surveillance plans, and performance evaluation reports. Quality system expectations reference ISO 13485 design controls and ongoing surveillance integration. For solo builders, this skill is advanced and niche: it does not replace notified bodies or qualified regulatory consultants, but it gives a structured checklist when you are shipping health-tech and need language that matches MDR annexes. Invoke it when classification, technical file structure, or clinical evaluation planning blocks your release timeline.735installs73Quality Manager Qms Iso13485Quality Manager QMS ISO 13485 is a specialized compliance skill for builders touching medical devices or SaMD who must speak auditor language, not just ship features. It walks ISO 13485:2016 clause by clause—QMS foundations, management responsibility, resources, product realization, and measurement—with explicit implementation notes, evidence expectations, and sample audit questions. Solo founders in health tech use it during Validate to decide whether their intended software scope fits a 13485-controlled path and what documented processes they need before engineering spikes. It also supports Ship and Operate when preparing design-history rationale, CAPA loops, and objective evidence for notified bodies or internal audits. The skill is reference-heavy rather than a codegen integration; your agent uses it to draft gap analyses, procedure outlines, and traceability language. It is not a substitute for a qualified regulatory consultant for classification or submission strategy.731installs74Capa OfficerCapa-officer is an agent skill that encodes corrective and preventive action (CAPA) effectiveness verification for quality teams shipping regulated or process-heavy products. Solo founders wearing a quality hat—or small teams without a full QA org—can use it to plan how they will prove a fix actually worked: timelines tied to severity, methods, criteria, closure rules, and what to do when a CAPA fails verification. The guide stresses planning verification before implementation, which prevents the common trap of closing tickets without data. It fits medical device, pharma-adjacent, and general ISO-style quality contexts where agents help draft plans and records. Pair it with your existing CAPA ticket or QMS workflow; the skill does not replace statutory submissions or auditor judgment but gives consistent structure for effectiveness evidence.730installs75Fda Consultant SpecialistFDA Consultant Specialist is an agent skill that acts like a regulatory consultant for medical device and SaMD teams. It helps solo builders and small manufacturers choose the right FDA pathway when a predicate exists or when the device is novel, then orient work toward 510(k) substantial equivalence narratives, PMA or De Novo when predicates fail, and the correct abbreviated versus traditional 510(k) flavor. Beyond pathway selection, it surfaces Quality System Regulation (21 CFR 820) expectations, HIPAA considerations when devices handle PHI, and FDA cybersecurity guidance that increasingly gates clearance. Use it when triggers mention 510(k), PMA, De Novo, QSR, predicate devices, premarket submissions, or FDA cybersecurity—not for general HIPAA SaaS unrelated to devices. Output is strategic regulatory structure; you still need qualified human RA/QA review before filing.726installs76Ms365 Tenant ManagerMS365 Tenant Manager is an agent skill for solo builders and small teams who need a structured first pass on a Microsoft 365 tenant without guessing admin center steps. It organizes work into five phases—initial configuration, custom domain, security baseline, service enablement, and compliance—with time estimates and priority labels so you can execute in one sitting or spread across a day. The skill surfaces concrete artifacts: DNS record counts and types, generated PowerShell setup scripts, and license mix guidance for executives versus line-of-business users. Use it when you are moving off consumer email, onboarding a new company domain, or rehearsing a clean tenant before production users land. It complements generic IT checklists by anchoring tasks to M365-specific dependencies (verification TXT, Exchange MX, Autodiscover) rather than abstract “enable security” bullets.726installs77Isms Audit Expertisms-audit-expert delivers a cloud security audit guide framed around the shared responsibility model and practical verification checklists for IaaS, PaaS, and SaaS deployments. Solo builders shipping on AWS, Azure, or integrated SaaS products can use it to structure reviews of identity integration, encryption, logging, network controls, and provider certifications without hiring a full audit firm for every release. The skill emphasizes what the customer still owns—data classification, app security on PaaS, access policies on SaaS—and maps audit focus to each layer. It suits security review workflows before go-live and when revisiting posture after architecture changes, though it is guidance rather than automated scanning.723installs78Information Security Manager Iso27001Information Security Manager ISO 27001 packages incident response know-how for solo founders and tiny teams who must behave like they have a security desk when something goes wrong. It organizes incidents into clear categories—breach, malware, data leakage, denial of service, policy violations, and physical events—and ties each to severity levels with explicit response clocks. Agents can draft escalation paths, fill communication templates, and walk recovery checklists without skipping ISO-minded documentation steps. Use it while you are hardening ship criteria, rehearsing tabletop scenarios, or operating live production when alerts fire. It is procedural guidance, not a certified ISMS audit replacement; pair it with your actual policies, tooling, and legal contacts.717installs79Quality Manager QmrQuality Manager QMR is an agent skill that guides ISO 13485 management reviews for medical-device and other regulated teams. It spells out Clause 5.6 expectations—planned intervals, top-management involvement, mandatory inputs and outputs, and maintained records—and pairs them with practical schedules from monthly operational updates through full semi-annual reviews. Solo builders rarely need a full QMS, but indie SaaS founders targeting FDA-style quality paths or selling into hospitals can use it to draft review agendas, input packets, and action logs without hiring a consultant for every cycle. The skill emphasizes documentation templates and traceable follow-up, which auditors expect. It is procedural reference material, not an automated connector; your agent structures content you already collect from CAPA, complaints, and internal audits.713installs80Mdr 745 Specialistmdr-745-specialist is a compliance-oriented agent skill that walks builders and consultants through EU Medical Device Regulation clinical evidence expectations under Annex XIV. It is aimed at teams shipping or scoping software as a medical device—not typical indie SaaS—who must decide how much clinical data they need before design freeze. The readme frames an evidence hierarchy, class-based minimums, and when randomized trials versus systematic literature may suffice. It connects clinical evaluation, literature strategies, investigation requirements, and post-market clinical follow-up into one checklist-style narrative so agents help draft structured thinking rather than casual regulatory guesses. Use it in Validate while scoping classification, claims, and equivalence: the output should inform risk files, clinical evaluation plans, and whether you must run investigations before Ship. Pair with human notified-body and regulatory counsel; the skill teaches process framing, not legal sign-off.711installs81Qms Audit ExpertQMS Audit Expert is an agent skill that encodes an ISO 13485:2016 internal audit playbook for medical device organizations. Solo technical founders in medtech—or tiny QA teams—use it when they must run Clause 8.2.4 programmes, prepare for certification or surveillance audits, or verify CAPA and design-control closure without guessing auditor expectations. The reference stresses design history files, IQ/OQ/PQ validation, controlled documents, corrective action loops, vigilance and customer feedback, and ISO 14971 risk alignment across the product lifecycle. It is prescriptive by design: unlike generic security audits, it maps directly to regulatory QMS evidence auditors expect for EU MDR and FDA QSR bridges.709installs82Sales EngineerSales Engineer is an agent skill packaged as a demo script template for indie SaaS founders and technical sellers who must run structured evaluation calls without a full sales ops team. It walks you through demo metadata, a practical pre-flight checklist (environments, VPN, integrations, recording), stakeholder mapping with interests, and a minute-by-minute agenda that balances account executive framing with sales-engineer depth. Use it when you are moving from prototype to paid conversations and need repeatable runbooks instead of improvised screen shares. The skill emphasizes backup environments, customer-specific branding when applicable, and explicit slots for core use cases plus integration proof. It does not provision infrastructure or close contracts; it produces a demo-ready script and checklist so Validate-stage conversations stay focused and recoverable if something breaks live.668installs83Campaign Analyticscampaign-analytics is a structured A/B test analysis template for solo marketers and founders who run experiments without a full experimentation platform write-up. The skill walks agents through naming tests, stating hypotheses in If-then-because form, locking design parameters like primary and secondary metrics, minimum sample size, minimum detectable effect, and confidence level, then capturing targeting and channel constraints before results land. It is ideal when you already have traffic and need a consistent doc for decisions—not for setting up tracking code. Use it during grow-phase optimization of landing pages, onboarding emails, or ad creative. Complexity is beginner-friendly because it is mostly fill-in-the-blank tables and narrative sections agents complete from your data. Deliverable is a decision-ready test report you can attach to backlog items or share with stakeholders.662installs84Jira ExpertJira Expert is a pattern library of Jira automation rules for solo and indie builders who still coordinate work in Jira while coding with agents. It walks through trigger–condition–action recipes you can paste into Jira’s rule builder: routing new issues to component leads or a round-robin queue, bouncing feedback waits back to reporters, keeping parent issues aligned with subtasks and epics, and paging Slack when critical bugs land. It does not call the Jira API from your repo; your agent helps you translate these examples into project-specific rules. Use it when you want less manual triage and fewer stale parent tickets without hiring a Jira admin. Pair with your issue templates and sprint rituals in Validate or Operate when backlog hygiene drifts.662installs85Scrum MasterScrum Master is an agent skill oriented around agile delivery analytics for small teams and solo builders who still want disciplined sprints. From the ingested output schema, it synthesizes multi-sprint velocity—totals, commitment consistency, and stability signals—then layers sprint health grades across dimensions like scope stability and blocker resolution. That combination helps you see whether you are consistently over-committing, whether scope creep is eating predictability, and whether blockers are dragging execution. Forecast bands give a pragmatic range for the next sprint instead of a single optimistic number, while anomaly callouts flag sprints that skewed the trend. Use it when you are self-managing delivery with AI agents and need an evidence-based nudge before planning, mid-sprint check-ins, or retros—without standing up Jira plugins or a data warehouse. It is methodology plus metrics, not a task tracker replacement.660installs86Revenue OperationsRevenue Operations is an agent skill oriented around B2B-style funnel analytics for founders who sell without a full RevOps team. Given pipeline snapshots, it reasons over coverage ratio against quota, stage conversion percentages, sales velocity components, and aging deals that exceed stage-specific day thresholds. Example output includes ratings such as At Risk when coverage sits near 2.2x versus a 3–4x target, plus enumerated aging deals with days over threshold. Solo builders running SaaS or services use it to translate CRM exports or JSON payloads into decisions: where leads stall, whether pipeline supports the quarter, and which opportunities need a push or kill. It complements validate-stage pricing conversations and grow-stage lifecycle work but does not replace a CRM—it interprets the numbers you already have. Pair with spreadsheet or CRM export skills upstream and with content or outbound skills when diagnostics show a weak mid-funnel.649installs87Senior PmSenior PM is a heavyweight project-management agent skill aimed at complex software and SaaS initiatives. For a solo builder, it is most valuable when a one-person shop still has investor updates, multiple features in flight, contractors, or compliance threads that need adult supervision on paper. The skill packages portfolio thinking—WSJF and related prioritization, expected monetary value on risks, Monte Carlo schedule uncertainty, capacity planning, and executive status narratives. It is not a lightweight todo helper; it assumes you are managing dependencies, budgets, and stakeholders at a scale closer to enterprise program management. Place it on the Build → pm shelf when you are turning an approved direction into a managed roadmap, then lean on it again in Ship and Operate for milestone and health reviews. Indie hackers shipping a weekend CRUD app can skip it; founders running a growing SaaS with real money and dates on the line get structured outputs instead of improvised Notion bullets.649installs88Confluence ExpertConfluence Expert is a documentation-focused agent skill for solo builders and tiny teams who treat Confluence as the system of record for specs, runbooks, and meeting notes. It walks through space creation, hierarchical page design, storage-format HTML bodies, versioned updates, CQL search, labels, and collaborative standards—primarily via the Confluence MCP server patterns documented in the skill. Use it when a wiki grew organically and needs restructuring, when you are standing up an engineering knowledge base with sane permissions, or when you want repeatable templates with macros instead of blank pages every sprint. It also supports audits and governance passes as the product matures. The skill is integration-oriented procedural knowledge for Atlassian Cloud workflows; you still need MCP credentials and org approval for write operations. It complements build-time feature work and later operate-phase runbook hygiene without replacing a technical writer for voice and tone.646installs89Customer Success Managercustomer-success-manager is an agent skill that packages a formal Executive Business Review (EBR) template for indie SaaS founders and solo operators who act as their own customer success lead. Rather than improvising renewal calls in chat, the skill supplies sectioned markdown for partnership metrics, strategic alignment, and quantified value delivered so an agent can populate real account data consistently. It emphasizes renewal timing, seat utilization, health scoring, and NPS alongside narrative alignment between customer business priorities and your product’s contribution. Use it in Grow when preparing QBRs or renewal conversations, and in Operate when you need a single source of truth before escalating at-risk accounts. The pattern is document generation from a proven CSM playbook—not live CRM API integration—so you bring the numbers and the agent shapes the executive story.646installs90Atlassian AdminAtlassian Admin is an agent skill that helps solo builders and tiny teams configure Jira-style project permission schemes without memorizing every Atlassian permission key. The bundled reference models a standard project permission scheme with versioned metadata, role definitions from project admins through read-only viewers, and explicit grants for administering projects, browsing issues, and viewing dev tools linked to commits and pull requests. Use it when you are standing up engineering workflow in Jira, reconciling who can transition issues versus comment, or onboarding contractors with viewer-only access. The skill is template- and policy-oriented rather than a live API integration—you shape JSON-aligned schemes and role-to-group mappings that map to dept-engineering, stakeholders, and org-wide groups. Beginner-friendly for builders who already use Jira but struggle with permission scheme sprawl. It supports Operate-phase iteration on access control as the product team grows, keeping documentation and agent output aligned with Atlassian’s permission vocabulary.624installs91Atlassian TemplatesAtlassian Templates is an agent skill that encodes a template governance framework for solo founders who scale into small teams on Confluence and Jira. Instead of one-off page templates drifting out of date, it defines concrete ownership: who reviews dashboards quarterly, who runs CQL adoption queries, and when a committee must approve cross-team templates. Thresholds are explicit—archive low-use templates, escalate changes affecting more than fifty users, and measure friction when sections get deleted. Stewards tag Jira feedback and link template pages so agent-assisted edits respect process. For Prism builders in Validate through Operate, it helps standardize specs, retros, and runbooks without reinventing governance each sprint. Use it when you are creating or refactoring Confluence templates, not when you only need a single ad-hoc doc with no maintenance plan.622installs92A11y Audita11y-audit is an agent skill that walks your frontend code for accessibility violations and explains what breaks WCAG-style expectations. Solo builders shipping customer-facing React or web UI install it when they want a second pair of eyes on contrast, keyboard use, semantics, and form labeling without hiring a full audit firm. Use it during late Build tweaks and during Ship review whenever you touch cards, tables, search bars, or action controls. The bundled samples mirror real mistakes—div-onClick handlers, unnamed images, color-only status, weak table structure—so the agent can pattern-match your repo. It matters because small indie teams often skip formal a11y work until a lawsuit, enterprise buyer, or App Store review forces the issue; catching AA failures early is cheaper than retrofitting. Pair it with your visual design pass and treat outputs as a fix list for the next coding session.606installs93Interview System Designerinterview-system-designer is an agent skill oriented around designing and operationalizing technical interview systems, with emphasis on system-design evaluation for senior engineering roles. The bundled examples model repeatable JSON scorecards: multiple interviewers, weighted competency areas such as system_design, technical_leadership, problem_solving, and explicit overall recommendations. For solo founders who must hire their first senior engineer without a dedicated recruiting org, the skill gives the agent a scaffold to define rubrics, normalize interviewer input, and reason about consistency across sessions. It is not a live ATS integration; it is procedural knowledge for how to structure interviews and capture comparable signals. Place it on validate/scope as the canonical shelf, but expect reuse in build/pm when defining team growth and in operate/iterate when refining hiring after early mis-hires.594installs94Content HumanizerContent-humanizer is an agent skill for solo founders and marketers who publish with LLMs but sound like every other AI blog. It assumes you have already stripped obvious AI tells and now need Mode 3: deliberate brand voice. The workflow starts by extracting a voice profile from one piece the brand is proud of—measuring sentence length, formality, humor, how the writer relates to the reader, and verbal tics—then applying structured voice types such as the Direct Expert. That makes edits repeatable across landing pages, emails, and social posts without re-explaining taste every session. It is most often shelved under grow/content because lifecycle and SEO content benefit longest from consistent voice, but it is equally useful in launch/distribution when shipping announcements or Product Hunt copy. Intermediate complexity: you need real brand examples, not invented personas. It is editorial workflow knowledge packaged as a skill, not an SEO keyword tool or plagiarism checker.592installs95Red Teamred-team is an advanced security methodology skill for solo builders and tiny teams who need structured offensive planning—not ad-hoc “try hacking prod” chat. It documents how to construct attack-path graphs aligned with MITRE ATT&CK: nodes for techniques and intermediate tactic states, edges for prerequisites, and weights from effort scoring that blends detection risk with dependency depth. Planners enumerate paths from realistic starting access (external, internal, credentialed) to crown jewels such as domain controllers or sensitive data stores, then prioritize choke points that appear on many routes. Indie use cases include pre-launch threat modeling for B2B SaaS, scoping paid pentests, or designing internal purple-team exercises without drowning in unstructured technique lists. The skill does not execute exploits; it produces analyzable graphs and scoring for decisions about controls, monitoring, and test depth. Treat outputs as planning artifacts subject to rules of engagement and legal authorization.584installs96Competitive Intelcompetitive-intel is an agent skill that walks solo builders through open-source competitive intelligence: where to look, what each source reveals, and how to interpret signals without a full analyst team. It catalogs free channels from pricing pages and review sites to hiring posts and ad libraries, plus optional paid stacks for serious rival tracking. Use it in Idea when picking a niche, in Validate when stress-testing positioning against incumbents, and again before Launch when sharpening differentiation. The skill is methodology-heavy rather than a single API integration—you bring competitor names and hypotheses; the agent applies the playbook to produce structured notes on positioning, roadmap hints, and customer pain. Outputs work best when saved as living docs you refresh quarterly rather than one-off chat dumps.576installs97Cold EmailCold Email is a deliverability-first agent skill for solo and indie builders who plan B2B outbound. It treats inbox placement as a stacked system: domain reputation, authentication, infrastructure, list hygiene, message content, and engagement—each must be correct before copy tweaks matter. The skill walks you through using a dedicated sending domain instead of your main brand domain, configuring SPF, DKIM, and DMARC, and respecting IP and volume ramp rules so blacklisting does not take down company email. It is written for founders selling SaaS, agencies, or services who are tempted to spray campaigns from acme.com on day one. Use it when you are drafting sequences but have not verified DNS, warmed a subdomain, or validated that list sources will not spike complaints. The outcome is a repeatable pre-flight checklist aligned with how inbox providers score trust, so later automation and personalization sit on stable rails rather than hidden spam traps.575installs98CopywritingCopywriting is an agent skill packaged as a Copy Frameworks Reference: headline formulas and page section types for solo builders who need credible marketing language fast. It is for founders writing their own landing pages, pricing blurbs, and campaign hooks when they do not have a dedicated copywriter. Use it when you are structuring a hero, reframing pain versus pleasure, or tailoring audience-specific value props for Shopify, teams, or self-serve SaaS. The reference groups patterns into outcome-focused, problem-focused, audience-focused, and differentiation-focused buckets with concrete examples so the agent can adapt tone without inventing a new framework each time. It does not replace brand strategy or full funnel analytics; it gives repeatable sentence molds and section logic you can iterate in Launch and reuse in Grow content. Intermediate complexity because you still need product truth and positioning; the skill supplies scaffolding, not claims you have not validated.574installs99Ad CreativeAd Creative is an agent skill that gives solo builders a working catalog of headline and copy frameworks for paid and organic distribution. It encodes a simple discipline up front: each ad should stop the right reader, deliver one clear message, and drive a single action—avoiding the common indie mistake of stacking three value props in one unit. Eight frameworks are indexed—from PAS and BAB through social proof, contrarian, and specificity angles—with PAS documented in depth including cold-audience fit, problem-agitate-solution rhythm, and a concrete SaaS stand-up chaos example with primary text and headline. Use it during launch when you are writing Meta, LinkedIn, or search creatives, and during grow when refreshing lifecycle or retargeting copy. It is intermediate editorial craft: you still need offer clarity and honest claims. Not for technical SEO page structure or ASO metadata—that is a different skill lane.573installs100Secrets Vault ManagerSecrets Vault Manager packages a cloud secret store reference for solo builders who must place API keys and database credentials in a managed vault instead of .env files in production. The skill compares AWS Secrets Manager, Azure Key Vault, and GCP Secret Manager across secret types, size limits, versioning, rotation hooks, encryption tiers, cross-region behavior, IAM or RBAC models, audit trails, and cost models so you can align with an already-chosen cloud rather than re-litigating vendors in chat. Use it while hardening Ship workflows and when Operate tasks revisit rotation or compliance. It is editorial reference material—agents should cite the matrices when recommending a provider, then follow your org’s Terraform or console runbooks to implement. Pair with deployment skills after you pick ARNs or vault URIs for ECS, EKS, Azure Functions, or Cloud Run bindings.572installs101Security Pen Testingsecurity-pen-testing packages a reference of safe, authorized attack-pattern probes—starting with cross-site scripting payloads for reflected contexts, filter evasion, and encoding tricks—that solo builders and small teams can use while shipping web SaaS, APIs with HTML admin UIs, or internal tools. It helps you systematically exercise inputs where user data echoes in responses, so you catch missing encoding before customers or bots do. The content assumes ethical scope: authorized assessments, labs, and defensive research only. Pair it with your own tooling (browser, proxy, scanner) during Ship; it does not replace a full pentest program or compliance audit but accelerates repeatable manual checks when you lack a dedicated AppSec teammate.570installs102Contract And Proposal WriterContract and Proposal Writer is a business-growth agent skill that produces jurisdiction-aware freelance agreements, project proposals, statements of work, NDAs, and master service agreements as structured Markdown. Solo builders and consultants use it when they must move from a verbal deal to written terms—especially across US, EU, UK, or DACH contexts where GDPR and local law clauses matter. The skill encodes tiered document types, clause references, and conversion guidance toward docx without pretending to replace an attorney on high-stakes deals. It fits Validate when you price and scope an engagement, and Grow when you repeat client packaging and renewal paperwork. Triggers align with drafting a freelance contract, preparing a client proposal, writing an SOW, or issuing an NDA before sharing IP.568installs103Cfo AdvisorCFO Advisor is a reference skill that walks solo and indie builders through startup cash management as the constraint that kills companies faster than margin. It explains the cash equation—beginning balance plus collections minus payroll, vendors, infra, debt, and financing—and stresses that revenue on the P&L is not the same as cash in the bank. The doc surfaces practical sources of float and capital builders often skip, from annual prepay and enterprise deposits to Net-60 terms and startup cloud credits, alongside sneaky outflows like prepaid SaaS, sponsorships, recruiter success fees, and fundraise legal bills. A worked scenario contrasts December accrual revenue on a million-dollar annual contract with when cash actually arrives, which is essential for forecasting without fooling yourself. Use it during pricing and scope conversations, before hires or big commits, and whenever you reconcile board-style metrics with bank balance. It does not replace accountants or tax counsel; it gives your agent structured vocabulary to stress-test plans and dashboards you already maintain.565installs104Sql Database AssistantSQL Database Assistant is procedural knowledge for solo builders who own both application code and database performance on Postgres-style workloads. It teaches how to run EXPLAIN (ANALYZE, BUFFERS), interpret planner nodes such as index scans versus sequential scans, and compare estimated rows to actual rows so you know when ANALYZE or index work is overdue. The skill connects observable symptoms—disk reads, sort spills, high loop counts—to concrete remediation rather than generic “add an index” advice. Use it when API endpoints lag, dashboards time out, or background jobs hammer the same tables. It fits indie SaaS and internal tools where you cannot afford a dedicated DBA but still need predictable query latency before launch and after traffic grows. Pair it with schema migrations and integration tests so index changes do not regress application behavior.565installs105Marketing OpsMarketing Ops is a central router and orchestrator for a larger marketing skill ecosystem, written for solo and indie builders who need a single entry point when they are unsure which marketing specialist skill to invoke. Instead of guessing between content, SEO, conversion, channel, and analytics playbooks, you describe the outcome or constraint and the skill identifies the right downstream skill or runs an ordered multi-skill campaign. It also supports a marketing audit mode that touches SEO, content, CRO, and channels in one pass. The workflow assumes optional marketing-context.md for grounding; without it, the skill steers you to establish context first so routing and priorities are not generic. Use it when marketing questions are broad, when you want a coordinated plan across skills, or when you need help sequencing work after validation and before or during growth—not as a replacement for deep specialist skills it routes to.562installs106Performance ProfilerPerformance Profiler is an agent skill for systematic performance engineering across Node.js, Python, and Go codebases. It pushes measure-first workflows: establish a baseline, profile CPU and memory, inspect frontend bundles, optimize database access, then verify with load tests. Solo builders use it when latency spikes before a release, memory creeps in production, dependencies inflate JavaScript bundles, or queries cross the hundred-millisecond line. The skill references concrete tooling—flamegraphs, py-spy, pprof, k6, Artillery—and a performance_profiler.py entry script for CI-friendly JSON risk scans. It is investigative and optimization-oriented rather than a monitoring dashboard replacement, fitting pre-ship hardening and post-launch iteration when users report slowness you cannot attribute to a single line of code.562installs107Content StrategyContent Strategy is a reference skill for solo builders who must ship marketing copy without a content team. It splits work into searchable pieces that answer existing queries and shareable pieces that spark conversation, and it tells you to prioritize search as the durable foundation. For searchable assets, the skill prescribes tight alignment to keyword intent, titles and headings that mirror how people query, keywords in title, H-tags, opening paragraph, and URL, plus thorough coverage with data, examples, and authority links—while also noting AI and LLM discovery through clear positioning and consistent brand signals. Shareable work should lead with insight, data, or contrarian reasoning, emotional story, or timely tension so readers pass it along. It introduces use-case content as persona-plus-job long-tail patterns such as project management for designers. Use it when planning a blog, docs hub, or landing copy during launch, when refreshing lifecycle emails in grow, or when turning validate-stage positioning into pages that rank. Pair with your analytics stack to see which pillar actually earns traffic.561installs108Skill Security Auditorskill-security-auditor is an agent skill that teaches a threat model for AI agent skills—how SKILL.md instructions, bundled scripts, and dependencies can be abused when users trust packages by default. Solo builders curating skills from marketplaces or authoring their own SKILL.md files can use it to reason about prompt injection, unsandboxed execution, filesystem persistence, and social-engineering tropes that make risky installs feel helpful. It does not replace automated scanners or Prism’s ingest security signals; it gives your agent a checklist mindset aligned with real attack surfaces called out in the doc. Canonical shelf is Ship/security because that is when you should block bad packages before production use, but the same analysis applies while integrating agent tooling during Build.561installs109Api Design ReviewerAPI Design Reviewer is a reference-style agent skill grounded in common REST API anti-patterns: treating endpoints as RPC verbs, mixing naming conventions in responses, and other choices that erode developer experience and maintainability. Solo and indie builders wiring backends for SaaS products, internal agents, or public integrations can invoke it when drafting OpenAPI specs, refactoring legacy `/getUser`-style routes, or reviewing a pull request that touches HTTP contracts. The skill does not replace a full security audit or performance test plan; it focuses on design hygiene so resources, methods, and payloads stay predictable for clients and codegen tools. Because poor API shape surfaces during implementation and again before launch, it fits both backend build work and ship-time review without locking you to a single toolchain—Claude Code, Cursor, or Codex can apply the same rules to your repo’s routes and DTOs.560installs110Landing Page GeneratorLanding-page-generator is an agent skill that turns a product description into complete Next.js/React TypeScript components styled with Tailwind. It ships multiple section variants—heroes, feature grids, pricing, FAQ, testimonials, CTAs, and footers—using PAS, AIDA, and BAB copy structures and SEO-friendly markup. Solo builders invoke it when they need a homepage, campaign page, lead capture surface, or A/B variants without hiring a copywriter first. Output emphasizes measurable performance budgets and schema where relevant, making it suitable for validation landings that later evolve into marketing sites at launch.560installs111Gcp Cloud ArchitectGCP Cloud Architect is a reference skill that helps solo and indie builders choose among six documented Google Cloud architecture patterns instead of guessing services ad hoc. It centers on a pattern selection matrix that compares serverless web apps, microservices on GKE, classic three-tier stacks, serverless data pipelines, ML platforms, and multi-region HA—each with indicative user scale, monthly cost ranges, and complexity. Builders shipping SaaS MVPs, mobile backends, e-commerce, analytics ETL, or MLOps on GCP use it when they need a coherent diagram-level blueprint before Terraform or console work. The content is pattern-oriented reference material rather than a deploy automation workflow, so it pairs well with separate IaC or console skills. Confidence is high for taxonomy placement as cloud infrastructure documentation aimed at build-time architecture decisions.559installs112Marketing PsychologyMarketing Psychology is a journey-wide agent skill that maps more than seventy mental models and behavioral principles to real marketing work—why people buy, where flows leak, and how to rewrite assets with evidence-backed persuasion. Solo builders invoke it when users mention psychology, cognitive bias, persuasion, or consumer behavior, or when a page simply will not convert. It runs in Diagnose mode to audit violations of behavioral principles, Apply mode to propose three to five concrete changes on a live asset, and Reference mode to explain a single model with examples. If marketing-context.md exists, the skill reads audience personas and positioning first. The canonical Prism shelf is Validate pricing because offer and objection psychology gates commitment before full build spend, but the same lens applies across landing copy in Launch, lifecycle email in Grow, and repositioning in Idea. It complements strategy docs; it does not replace analytics or legal compliance review.558installs113Codebase Onboardingcodebase-onboarding is a documentation template skill that guides an agent through drafting a production-style README for an existing or in-progress repository. Indie builders install it when the code works locally but the repo still fails the “clone in five minutes” test that collaborators, buyers, or future-you will need. The embedded structure covers one-line positioning, CI and coverage badges, environment URLs, prerequisite matrix, dockerized dependencies, migration and seed commands, and a short verification checklist so setup failures surface early. It also prompts an architecture overview with ASCII flow and a stack table explaining technology choices. Use it in Build when formalizing docs, after a validate-phase prototype graduates to a real repo, or in Operate when revisiting runbooks—without replacing deep ADRs or auto-generated API reference.557installs114Incident ResponseIncident Response is an agent skill focused on regulatory notification deadlines and disciplined declaration timing when personal data, cardholder data, or PHI may be involved. Solo founders and tiny teams without a dedicated GRC function use it at the moment an incident is declared to avoid missing supervisory authority, card brand, or HHS clocks that start immediately—not after forensics completes. The embedded reference table contrasts GDPR, PCI-DSS, HIPAA, and related frameworks on jurisdiction, incident type, deadline, recipient, and penalties, with an explicit rule to assume the strictest applicable timeline when scope is still unknown and record how that assumption was resolved. Primary placement is operate errors because it governs live breach response, while ship security remains relevant when baking notification playbooks into launch readiness. It does not replace legal counsel but gives agents citable, structured facts for runbooks, tickets, and first-hour communications.557installs115Ci Cd Pipeline Builderci-cd-pipeline-builder helps solo and indie developers who have a working codebase but no trustworthy automation yet. Two Python scripts form the core: one inspects the repository and writes a structured stack profile, the other turns that profile into a practical CI workflow for GitHub Actions or GitLab CI. The skill encodes opinionated deployment gates—lint must pass before tests, tests before builds, artifacts before deploys, and production promotion behind manual approval on protected branches—so you start from a sane baseline instead of copy-pasting fragile YAML. It ships reference material for platform-specific templates and gate policies, and documents install paths for Claude Code, Codex, and OpenClaw skill directories. Use it when moving from local-only builds to staged delivery; extend the generated files with your secrets, runners, and cloud-specific deploy steps rather than treating output as production-final without review.554installs116Content ProductionAI Citation Readiness is an agent skill that teaches solo builders how to format web content so AI answer engines cite their pages. It targets the emerging AEO layer—visibility inside generated answers—not only traditional search rankings. The skill documents five citation signals, with emphasis on answer-first paragraphs immediately under each H2 and passage-level chunks of roughly 120–180 words that remain coherent when quoted out of context. Builders shipping landing pages, docs, or blog posts with Claude Code, Cursor, or Codex use it during launch and growth passes when they want distribution through AI referrals. The readme includes before/after examples (e.g., framework choice copy) so agents can rewrite buildup-style prose into claim-first, evidence-backed openings. It suits indie SaaS founders and content-led products who own their own marketing copy and need repeatable rules rather than one-off prompts.552installs117Llm Wikillm-wiki is a journey-wide agent skill that teaches an LLM to act as maintainer of a personal or team research vault. Sources land in raw/ and stay immutable; the agent builds knowledge in wiki/ using a fixed folder schema—index catalog, timeline log, entities, concepts, per-source summaries, comparisons, and synthesis pages—with mandatory YAML frontmatter on every page. AGENTS.md (or CLAUDE.md) defines the evolving schema alongside the user. Solo builders use it when competitor articles, papers, and notes pile up and chat context is not enough to retain relationships between people, products, and ideas. It fits early Idea research, Validate scoping when you need comparable frameworks, Build when agents need durable context, and Grow when content strategy depends on prior synthesis. The pattern favors progressive curation over one-shot summaries and keeps provenance tied to source counts in frontmatter.551installs118Adversarial ReviewerAdversarial Reviewer is a prompt-only agent skill that breaks the self-review monoculture by forcing three hostile personas to scrutinize your changes. Solo builders install it when they are about to merge a PR, have staged or unstaged edits, or suspect the model is being too agreeable about code quality. The Saboteur hunts production breakage, the New Hire stresses maintainability and clarity, and the Security Auditor applies OWASP-informed pressure. Every persona must report at least one issue, and overlapping findings get promoted one severity level so consensus signals real risk. Output is severity-classified with explicit BLOCK, CONCERNS, or CLEAN verdicts rather than vague praise. There are no external tool dependencies—invoke it as a structured review ritual on recent changes. It complements ordinary linters and happy-path review skills by optimizing for blind spots, not style nits alone.550installs119Docker Developmentdocker-development is an opinionated agent skill for solo builders shipping services in containers. It activates when you ask to optimize a Dockerfile, improve docker-compose, implement multi-stage builds, audit container security, shrink images, or follow hardened container patterns. The skill routes through three slash commands: optimize for size and cache efficiency, compose for orchestrated local and deploy stacks, and security for Dockerfile and runtime review. It emphasizes build performance, layer caching, secret handling, and patterns that reduce wasted space and exposure—aimed at engineers who already use Docker but want agent-guided refactors rather than documentation spelunking. Pair it with your backend or API codebase when you need reproducible environments before Ship-phase CI hooks. It assumes you have existing Docker assets to analyze rather than teaching containers from zero.549installs120Code To PrdCode-to-prd is an agent skill for solo and indie builders who already have code in git and need a honest product picture without manually clicking every route. The workflow ingests project metadata and directory signals, then outputs a machine-readable snapshot: frameworks such as Next and React, dependency versions, counts of components and modules, and enumerated frontend paths alongside HTTP methods on backend handlers. That artifact acts like a draft PRD backbone—what ships today, not what you wish you had—so you can validate scope with cofounders, agents, or clients before committing to the next build sprint. It is strongest when the repo is the source of truth and weakest when requirements live only in slides or tickets. Pair it with planning or docs skills after the scan so narrative requirements and implementation tasks stay aligned with what the scanner actually found.548installs121Page Cropage-cro packages conversion_audit.py, a Python command-line checker that scans HTML for conversion-rate optimization signals solo founders care about on landing pages. You point it at a saved page or a public URL and it walks the DOM with a lightweight parser to inventory primary CTAs, form complexity, social-proof cues, trust mentions, and whether a mobile viewport meta tag exists. It is built for indie builders validating a landing page before paid traffic, refining a launch page, or revisiting lifecycle pages without subscribing to a full analytics suite. The workflow is deliberately simple: run the script, read structured findings or JSON, then fix gaps in copy, layout, or trust blocks. It does not replace qualitative user research or A/B testing platforms, but it gives a reproducible baseline audit your coding agent can run in CI or before every deploy.548installs122Database Schema Designerdatabase-schema-designer is an agent skill that walks solo and indie builders through production-grade relational design using Prisma on PostgreSQL. It centers on a complete reference schema for a multi-tenant task-management SaaS: organizations, users, membership roles, projects, and audit logging with created/updated/deleted timestamps and explicit table maps. Install it when you are greenfielding a backend or refactoring tables before migrations, and you want the agent to propose normalized entities, tenancy boundaries, foreign-key delete behavior, and indexing that match how you will query in production—not one-off demo tables. The skill complements ad-hoc chat schema questions by anchoring answers to a realistic SaaS shape you can adapt for your own product, reducing rework when APIs and auth rules depend on stable IDs and org isolation.547installs123Dependency Auditordependency-auditor is an agent skill for scrutinizing Go dependency manifests—especially rich `go.mod` files with deep indirect trees like Gin, testify, pgx, and uber/zap. Solo builders shipping small APIs or internal tools use it when a dependabot bump lands dozens of transitive modules and you need a structured pass over what actually entered the graph, which versions are pinned, and where known-risky families (crypto, JWT, database drivers) concentrate. The skill orients the agent around module boundaries and indirect comments rather than rewriting application code. It is best invoked after dependency upgrades or before production cutovers when you cannot afford a silent vulnerable indirect. Catalog listing aligns the skill with Security and Ship workflows even though the artifact is a backend go.mod; outcomes are review notes, upgrade recommendations, and a prioritized follow-up list rather than automated CVE database guarantees unless your agent workflow adds those tools.547installs124Threat DetectionThreat Detection is an advanced security agent skill for finding activity that slipped past automated controls. It combines structured hunting hypotheses, IOC sweep workflows, statistical anomaly detection (including z-score approaches), and MITRE ATT&CK–aligned signal prioritization so solo operators and small teams can turn noisy telemetry into ranked leads. The skill explicitly separates proactive detection from incident response (containment and recovery) and from red-team emulation. Workflows cover deception and honeypot hooks where you already instrument them. Indie builders running SaaS or internal APIs use it when they have logs or EDR data and need a repeatable hunt plan—not ad-hoc grep sessions. Outputs emphasize analyzable signals, hunt steps, and prioritization rationale rather than declaring breaches without evidence.546installs125Brand GuidelinesBrand Guidelines is a marketing identity skill that helps solo builders and tiny teams enforce or define visual and verbal standards across products and campaigns. When someone mentions brand colors, typography, logo usage, tone of voice, or company design standards, the agent acts as a brand identity expert: read optional product marketing context, decide whether to apply an existing system or document a new one, and walk through color systems, typography, logo rules, imagery guidance, and a tone matrix. Anthropic-specific work can lean on the bundled Anthropic identity section; other companies use the framework to assess gaps and write usable standards. It is procedural rather than a design tool—you still need your assets and decisions—but it reduces drift between landing pages, in-app UI copy, and launch assets. Use when consistency matters more than one-off creative experiments, especially before scaling content or paid distribution.545installs126Experiment DesignerExperiment Designer encodes a practical experimentation playbook for solo builders and small teams who must decide whether a change truly improved outcomes—not just spiked short-term clicks. It distinguishes A/B, multivariate, and holdout designs, then forces clarity on a single primary metric aligned to user value, plus guardrails that block harmful local wins and diagnostics that explain mechanism without becoming accidental gates. Stopping rules are defined upfront to curb peeking, mid-test metric swaps, and retroactive segmentation without correction. The skill also calls out novelty and primacy biases and suggests longer runs and cohort checks when stakes are high. It fits indie SaaS, ecommerce, and content products transitioning from gut feel to measurable iteration. Use it when you have enough traffic to test and need a written experiment spec your agent or team can execute and audit later.545installs127Seo AuditSEO Audit is a checker-oriented agent skill that gives solo builders and small teams a precise Core Web Vitals reference for technical SEO health scoring—especially when paired with scripts like seo_health_scorer.py in the same skill family. It codifies 2026 thresholds for Largest Contentful Paint, Cumulative Layout Shift, and Interaction to Next Paint, then ties each band to pass, warn, or fail outcomes so agents do not hand-wave “make it faster.” Fix guidance spans image optimization, server TTFB, layout stability, third-party script impact, and main-thread task splitting. Use it at Launch when pages are live and you need defensible SEO performance gates before scaling distribution or paid acquisition. It complements content and keyword work by attacking the measurable UX layer Google uses in ranking systems. Lab tools and CrUX field data are both named so you know which signal you are judging.545installs128Change ManagementChange Management is an agent skill that packages a startup-speed ADKAR playbook so solo and indie builders can announce and land organizational changes without chaos. It targets the gap where most changes fail—not in design but in implementation—by sequencing Awareness (why), Desire, Knowledge, Ability, and Reinforcement instead of leading with what is changing. The skill includes communication templates, common resistance patterns, and guidance for change fatigue when you stack tool migrations, strategy pivots, restructures, or product sunsets. It is written for founders and small teams who are effectively their own HR and ops: use it when you mention change management, rollout, managing resistance, org change, reorg, or pivot communication. Outputs are structured rollout steps and messaging you can adapt before you flip switches in production or in your own workflow.544installs129Chief Of StaffChief of Staff is a meta-advisory skill that acts as a router for founder questions: it matches what you asked to the best executive perspective using a detailed scoring matrix, then decides whether one role, two roles, or a synthetic board conversation should answer. Finance prompts (runway, burn multiple, bridge vs cuts, round terms, 18-month models) lean CFO; people and culture crises lean CHRO; product strategy leans CPO; revenue and pricing blends CRO and CFO. Solo builders wear every hat, so the skill helps you not treat a pricing question like a pure engineering task—or a termination like a spreadsheet exercise. It is journey-wide because capital, hiring, product, and GTM questions surface in Validate through Operate. Invoke when you have a strategic fork and want structured multi-lens reasoning instead of a single generic chat reply. It does not replace lawyers, accountants, or live metrics—you still supply real numbers.544installs130Env Secrets Managerenv-secrets-manager equips coding agents to treat environment variables and credentials as a controlled surface—not an afterthought. It catalogs high-signal secret patterns (provider API keys, GitHub PATs, AWS identifiers, Slack tokens, PEM blocks, and common assignment smells), assigns severity with explicit rotation guidance, and walks through a concise incident response sequence when something leaks. Solo builders install it while hardening repos before first deploy and again when onboarding new services: commit only examples, keep gitignore aligned, prefer managed secrets in staging and production, and run a validate-env script so DATABASE_URL and APP_SECRET cannot be empty at boot. The skill is procedural security knowledge rather than a hosted scanner—ideal when you want the agent to reason about exposure, suggest fixes, and scaffold checks that fit Claude Code or Cursor workflows without replacing dedicated secret-scanning products.544installs131Product DiscoveryProduct Discovery is an agent skill for solo and indie builders who need a disciplined alternative to jumping straight from a hunch into a backlog. It guides you to set one measurable outcome, build an Opportunity Solution Tree grounded in user evidence, map and prioritize assumptions across desirability, viability, feasibility, and usability, then validate the problem and solution with interviews, prototypes, and experiments that measure behavior—not slogans. A bundled assumption_mapper.py script supports scoring assumptions from CSV input, and the workflow embeds short discovery sprints with daily evidence review so weak opportunities get rejected early. Use it when you are validating product opportunities, planning discovery work, or testing fit before allocating build resources. The output is decision-ready clarity: which bets to pursue, what to prototype next, and what evidence you still lack—without replacing legal, financial, or market-size due diligence outside your control.544installs132Social ContentSocial-content is a platform-specific strategy guide for solo and indie builders who ship products and need consistent organic distribution without a marketing team. It consolidates practical rules for major networks—starting with LinkedIn and Twitter/X—including audience fit, posting frequency, timing windows, content formats that earn reach, and explicit anti-patterns that suppress distribution. The skill is meant when you are planning a content calendar, rewriting a launch thread, or tuning thought-leadership posts for B2B or tech audiences. It does not replace analytics tooling or ad buying; it gives procedural editorial knowledge so your agent drafts posts that match how each platform’s feed actually behaves. Use it alongside validate or launch work when you need messaging discipline, then keep it in rotation during grow for weekly publishing habits.543installs133Agent DesignerAgent Designer packages an Agent Evaluator workflow: feed structured execution logs from multi-agent runs and get quantitative performance analysis aimed at solo builders operating Claude Code, Codex, or custom agent stacks at scale. Each log entry carries task metadata, timing, token and USD cost fields, status including partial and timeout outcomes, action traces, tools used, and optional error details. The skill drives a Python analysis path that aggregates success rates, cost per task, latency statistics, and counters for failure modes, then ties those metrics to bottleneck hints and improvement opportunities rather than vanity dashboards. Use it when agents feel expensive or flaky and you need evidence—which task types fail, which tools dominate time, where retries cluster—before rewriting prompts or splitting agents. Complexity is intermediate because you must export consistent JSON logs from your runtime. It complements design-time architecture skills by closing the loop with empirical evals; it does not replace security review of agent permissions or external API keys.542installs134Product Analyticsproduct-analytics is an agent skill that gives solo builders ready-made dashboard blueprints instead of blank spreadsheets. It defines three complementary views—executive leadership signal, full-journey product health, and feature launch quality—each with section lists, suggested KPI blocks, and visualization hints such as cohort heatmaps and funnel waterfalls. You invoke it when you need to translate vague “track retention” goals into concrete metrics like activation rate, W8 retention, paid conversion, first-use adoption, and repeat usage tiers. The skill is template-driven planning artifacts for analytics stacks (whatever warehouse or BI you use), not a specific vendor integration. It also supports earlier journey moments when you need a metric frame for validation or launch reviews, but its home is Grow analytics where compounding and bottleneck detection matter.542installs135Rag ArchitectRAG Architect is an agent skill that guides solo builders through designing, implementing, and tuning retrieval-augmented generation systems. It spans document ingestion and chunking (fixed character or token windows with 10–20% overlap, sentence and paragraph-aware splits), embedding and index choices, vector search behavior, and evaluation so answers stay grounded in your corpus. Use it when you are adding “chat with my docs” to a SaaS, building an internal agent, or replacing naive full-document prompts with scalable retrieval. The skill emphasizes tradeoffs—predictable chunk sizes versus semantic boundaries—and production concerns like latency and accuracy. It fits the build phase for agent-tooling but also informs validate prototypes and later operate iterations when retrieval quality drifts. Pair it with your stack’s vector DB and observability once the pipeline is live.542installs136X Twitter GrowthX Twitter Growth is a marketing-focused agent skill that helps solo builders and founders treat X as a growth engine—not a random posting channel. It opens with a profile audit (bio, niche, pinned narrative) and continues into X-native tactics: how threads are structured for retention, how replies build distribution, and how to research competitors on the platform. It complements broader skills like social-content and social-media-manager by going deep on algorithm behavior and engagement engineering rather than generic captions. Use when you want to grow on X, write tweets or threads, analyze your profile, plan posting strategy, or optimize engagement. Ideal once you have a product story or launch moment to amplify across validate, launch, and grow phases.542installs137Changelog GeneratorChangelog Generator automates release notes from Conventional Commits with Keep a Changelog output and strict commit linting for CI-friendly release workflows. Indie builders maintaining a SaaS repo, CLI tool, or API surface install it when tag-based releases feel manual or when commit messages drift from Conventional Commits and break semver inference. The skill centers on two Python scripts: generate_changelog.py for --from-tag/--to-tag ranges with --next-version and markdown or JSON formats, and commit_linter.py for --from-ref/--to-ref linting with --strict mode before merge or release. Editorial rules mirror Keep a Changelog: security first, one user-visible bullet per change, impact-led wording, and explicit migration notes when behavior breaks. Reference docs cover CI integration, changelog formatting, and monorepo tagging strategies so you can wire the same commands into GitHub Actions or similar without reinventing release hygiene. Complexity is beginner to intermediate depending on monorepo layout; the skill is a procedural package you copy into Claude Code, Codex, or OpenClaw skill directories rather than a hosted service.541installs138Cmo AdvisorCMO Advisor is an agent skill that packages chief marketing officer–level judgment for solo builders and small teams scaling a product company. It is meant when you are choosing how the business should grow—product-led versus sales-led versus community-led—how brand and category positioning should read in the market, how to split marketing budget across channels, and how to shape a marketing organization as pipeline expectations rise. The skill explicitly defers execution work such as campaigns and content creation to other skills; this one is the strategic engine backed by procedural frameworks for brand positioning, growth frameworks, and marketing org design. Quick start paths invoke Python helpers to model budget allocation with MQL output by scenario and to simulate MRR growth under different growth models, so conversations can stay anchored in payback, CAC, LTV, and channel mix instead of generic advice. Use it when mentions of CMO, growth loops, marketing ROI, or pipeline coverage appear in planning, not when you only need copy or channel tactics.541installs139Google Workspace CliGoogle-workspace-cli is an agent skill that documents the gws command-line interface through ten role-based persona profiles, each bundling recipes for Gmail, Calendar, Google Tasks, Sheets, and related helpers. Solo builders who live in Google Workspace can install this so Claude Code or Cursor proposes the right gws invocations for morning briefings, meeting prep, standup reports, task creation, and project status pulls—without re-reading OAuth and API docs every time. It spans operating your business inbox and calendar while also supporting build-phase PM rituals like sprint scheduling and tracker reads. The skill is integration-heavy: you need the gws CLI authenticated in your environment. Treat personas as scoping lenses—the Executive Assistant and Project Manager flows are the most common for one-person teams wearing every hat.541installs140Cloud SecurityCloud Security is an agent skill that packages cloud security posture management (CSPM) as actionable check matrices for AWS, Azure, and GCP. It is built for solo and indie builders who deploy on public cloud and need consistent IAM and posture reviews without guessing which misconfigurations matter. Each row ties a concrete finding condition—such as Action and Resource wildcards, Principal: *, or risky iam:PassRole pairings—to severity, a MITRE ATT&CK mapping, and remediation steps you can hand to an agent or implement yourself. Use it when you are drafting policies during backend work, gatekeeping a release, or triaging posture drift in production. The skill complements automated scanners by giving your coding agent the same vocabulary security teams use, so reviews stay repeatable across providers instead of one-off chat advice.540installs141Mcp Server BuilderMCP Server Builder is an agent skill package for solo and indie builders who already have—or can export—an OpenAPI contract and want a repeatable path to Model Context Protocol tooling instead of hand-rolling every tool schema. It centers on two scripts: one that turns API operations into a manifest and language-specific scaffold, and one that enforces quality on those definitions before agents depend on them. The workflow matches how small teams ship fast without skipping schema discipline—generate, validate in strict mode, then iterate on the scaffold using the included Python and TypeScript reference templates. You reach for it during Build when Claude Code or similar agents need first-class access to your product API, especially when operation counts grow and naming consistency matters. Production focus means validation is treated as a gate, not an afterthought, which reduces broken tool calls and ambiguous parameters in live agent sessions.540installs142Observability DesignerObservability Designer is an agent skill that helps solo builders and tiny teams shape production monitoring before incidents become guesswork. It surfaces concrete alert patterns—latency percentiles, availability, and error budgets—with labels, annotations, and runbook hooks that match how real on-call workflows run. Use it when you are hardening a shipped API or SaaS and need PromQL-flavored rules that are reviewable in chat, not a black-box SaaS dashboard only. The skill emphasizes designing alerts with severity, ownership, and historical context fields so you can tune noise versus signal. It pairs naturally with operate-phase monitoring and infra work; it is not a full incident command system or hosted metrics backend. Indie operators who self-host or use managed Prometheus stacks benefit most when they want citable alert JSON and naming consistency across services.540installs143Saas Metrics CoachSaaS Metrics Coach is an agent skill for solo and indie builders running subscription products who need a fast, honest read on business health without spinning up a full analytics stack. You fill a short template—market segment, ARR stage, current and prior MRR, expansion and churned MRR, customer adds and losses, gross margin, and optional profit margin—and the skill coaches your numbers against published industry benchmark tables for churn and LTV:CAC, with explicit CRITICAL, WATCH, and HEALTHY thresholds that change by ACV band. It is built for moments when you are wondering whether churn is normal, if you are burning cash on acquisition, or how your stage compares to OpenView, Bessemer, SaaS Capital, and Paddle reference ranges. Use it during growth reviews, before pricing or packaging changes, or when you need a written narrative for investors or your own roadmap. The skill emphasizes interpretation and prioritization, not live data plumbing.540installs144Browser AutomationBrowser-automation documents anti-detection patterns for Playwright so solo builders can automate real browser workflows—E2E tests, onboarding flows, data collection—without tripping obvious bot signals. It organizes what anti-bot systems inspect into tiered detection vectors from trivial headless tells through advanced fingerprinting and behavioral analysis, then lists complementary stealth mitigations. The skill assumes you already chose browser automation and need engineering tactics, not a product strategy. It fits Build when integrating scrapers, agent browser tools, or CI browsers that must resemble human sessions. Respect site terms, robots policies, and legal constraints; the reference is defensive engineering for legitimate automation you own or are authorized to test.539installs145Pr Review ExpertPR Review Expert is an agent skill for structured review of GitHub pull requests and GitLab merge requests. Solo builders and small teams use it before merging changes that touch shared libraries, public APIs, database schema, or security-sensitive paths, especially when diffs exceed a few hundred lines or onboarding contributors need consistent feedback. The workflow goes beyond style: blast radius tracing, security-focused scanning, test coverage delta, breaking-change detection, ticket scope checks, and performance red flags such as N+1 queries or bundle regressions. It expects you to fetch the real diff from your forge. Output is a reviewer-ready report with a large checklist and prioritized findings you can paste into review comments or use as a merge gate. Pair it with your CI and dedicated security tooling; this skill organizes human-grade review judgment, not automated policy enforcement.539installs146Agent Workflow DesignerAgent Workflow Designer is a template skill for solo and indie builders who are stitching together multiple agents instead of one bloated system prompt. It documents five reusable workflow patterns—sequential pipelines, parallel fan-out/fan-in, intent routers with fallbacks, orchestrators with DAG-style dependencies, and generator–evaluator loops with iteration caps and score thresholds—each as copyable JSON so you can align your implementation with a named structure before coding. Selection heuristics spell out when linear order matters, when to trade latency for parallelism, when specialized handlers beat a generalist, and when quality gates must block finalization. The handoff minimum contract forces every step transition to carry identity, task text, constraints, upstream artifacts, and a token budget, which reduces dropped context between specialists. Use it while scoping voice squads, research pipelines, or shipping automations where handoffs are the failure mode. It does not call external APIs; you apply the patterns in your repo or orchestration layer.538installs147Analytics TrackingAnalytics Tracking is an agent skill for solo builders who need GA4 and Google Tag Manager wired correctly—not a pile of conflicting events. It guides you from a blank tracking plan through event naming, conversion mapping, custom dimensions, and UTM discipline, and it doubles as an audit playbook when events disappear or numbers disagree with reality. The skill treats broken or duplicated instrumentation as a product risk: you get checklists and debugging flows aimed at consented, parameterized events you can actually use for validation and growth. It explicitly is not for analyzing live ad campaigns (use campaign-oriented skills) or for deep in-product behavioral analysis (use product-analytics-style skills). Best when you are launching something new, migrating tags, or suspect your funnel metrics are lying to you.537installs148Playwright ProPlaywright-pro is a focused agent skill that acts as an assertions reference for solo builders shipping web apps with Playwright. It catalogs web-first expectations—visibility, text, values, attributes, accessibility-related state, counts, CSS, and screenshot baselines—alongside page-level URL and title checks, all using Playwright’s auto-retry semantics so tests survive slow networks and reactive frameworks. The skill emphasizes what to always use versus anti-patterns you should never copy, which is exactly the guidance indie teams need when one flaky suite blocks launches. Install it when you or your coding agent are authoring or refactoring e2e specs in TypeScript and want consistent, quotable patterns instead of reinventing assertion syntax per file. It does not replace a full test strategy, CI wiring, or fixture design, but it tightens the highest-friction part of UI automation: deterministic waits expressed as assertions. Complexity is beginner-friendly for anyone who already runs Playwright locally.537installs149Api Test Suite Builderapi-test-suite-builder is an agent skill for solo builders shipping HTTP APIs who want copy-paste-quality test files instead of improvised fetch checks. It documents end-to-end examples using Vitest and Supertest against application servers, including lifecycle hooks that create users, mint valid and expired JWTs, and tear down data after runs. The skill emphasizes security-relevant cases—unauthenticated access, malformed tokens, and expired credentials—alongside happy-path coverage so agents do not stop at a single 200 assertion. It fits indie SaaS and Next.js API routes but generalizes to any Node test runner pairing with Supertest. Use it when a route exists and you need a disciplined suite before CI or production traffic, not when you are still sketching OpenAPI-only mocks without a runnable app.536installs150Ciso AdvisorCISO-advisor is a compliance roadmap skill that helps solo founders and small SaaS teams decide which security and privacy frameworks to tackle first instead of boiling the ocean. It encodes a practical sequencing lens: who your customers are determines whether SOC 2 Type II, HIPAA, GDPR, ISO 27001, or government programs like FedRAMP belong on the near-term path. The reference material walks through what SOC 2 attestation means, which AICPA Trust Service Criteria to add beyond mandatory Security, and how Type I and Type II timelines differ in buyer credibility. Builders typically invoke it while scoping enterprise deals, before audit spend, or when EU residency forces GDPR alongside commercial certifications. It does not replace lawyers or auditors; it gives your agent structured vocabulary to draft roadmaps, backlog controls, and sales-facing security narratives aligned with how indie B2B products actually close deals.536installs151Competitive TeardownCompetitive Teardown gives solo builders ready-made competitive analysis scaffolds instead of blank docs when researching a market. The skill centers on a SWOT grid that forces evidence and impact ratings for internal strengths and weaknesses, plus external opportunities and threats with mitigation hooks. It also includes Porter's Five Forces applied to your product category—rivalry, switching costs, differentiation, and growth rate—so you can judge whether a niche is structurally attractive before you validate or build. Use it when you are mapping direct competitors, preparing positioning for a landing page, or updating strategy after a rival ships a major feature. The templates are agent-friendly tables your coding assistant can populate from public sources you provide. Outputs are decision inputs: where to differentiate, which weaknesses to exploit, and which threats need a concrete response plan—not vanity slides.536installs152Incident CommanderIncident Commander is an agent skill structured as a fill-in incident report template for solo operators and small teams who cannot afford a full SRE comms team during outages. It walks you through executive summary, quantified impact, customer-visible symptoms, a phased UTC timeline, and documented decision points—aligned with how incident commanders communicate while severity is still elevated. Use it in Operate when alerts have already fired and you need a single artifact stakeholders can read without joining the war room. It does not replace paging, runbooks, or root-cause tooling; it standardizes the narrative so postmortems and customer updates start from the same skeleton. Intermediate complexity because you must supply real metrics and honest scope; the skill keeps sections consistent so you ship the report faster under time pressure.536installs153Stripe Integration ExpertStripe Integration Expert is an agent skill for solo and indie builders who need production-grade billing instead of copy-pasted snippets that break under upgrades, retries, or webhook duplicates. It walks through subscription lifecycle management including trials, proration, and usage-based pricing, plus one-time checkout flows, customer portal access, and invoice handling. The skill emphasizes reliability: verified webhooks, idempotent event processing, and local replay with the Stripe CLI so you can debug delivery before launch. Use it when integrating Stripe for the first time, fixing flaky webhook handlers, migrating from another processor, or adding metered billing to an existing plan matrix. It maps common implementation choices to Next.js, Express, and Django so your agent can align routes, secrets, and database state with Stripe objects. The outcome is a billing layer you can ship with fewer revenue-impacting edge-case surprises.536installs154Ai SeoAI SEO is a reference skill for solo and indie builders who need their sites chosen before an LLM synthesizes an answer. It explains how major AI search platforms broadly index, retrieve passages, and cite sources, and where you can still influence selection upstream of generation. The material is oriented toward practical leverage—getting crawled, retrieved, and quoted—rather than generic SEO slogans. Google AI Overviews receives detailed treatment: you generally must already rank in the traditional top ten, favor structured schemas, and place crisp definitional paragraphs where extractors can lift them. The skill fits founders shipping landing pages, docs, or content-led SaaS who are optimizing for Perplexity-style citations and AI overview boxes, not only blue-link rankings. Treat the dated landscape note as a reminder to re-test behavior on each platform because retrieval policies shift faster than classic algorithm updates.535installs155Copy EditingCopy Editing is an agent skill that helps solo builders polish marketing and product copy so it reads human rather than model-flat. It explains how burstiness—the variance of sentence lengths—flags overly uniform AI rhythm, with clear CV bands from natural through strong AI signals. It also covers vocabulary diversity via type-token ratio across sliding windows, where repetitive safe word choices inflate AI likelihood. For each flagged pattern, the skill gives concrete rewrite moves: insert short sentences, split long lines, allow fragments, and alternate paragraph sizes. It is framed as reference material for an AI content detector workflow, so your agent can review drafts before lifecycle emails, blog posts, or landing hero sections go live. Use it when generated copy sounds generic, when you worry about detection or reader trust, or when you want consistent editorial pass rules the agent can apply pass after pass.535installs156Founder Coachfounder-coach packages practical self-management tools for solo founders who act as their own CEO. The centerpiece is a Weekly CEO Reflection you complete in about fifteen minutes every Friday: it forces clarity on your highest-leverage contribution, where you added little value, what you should have delegated, decisions you are avoiding, and one specific redo for the past week, then sets a single outcome and guardrail for the week ahead. A companion Energy Audit asks you to time-block a full week and label activities by energy impact so you can redesign schedules around deep work instead of reactive busyness. The skill does not replace a co-founder or coach chat; it gives repeatable templates that compound leadership judgment across Idea through Operate. Use it whenever you feel over-involved in details, stuck on a hard decision, or unable to name what actually moved the business forward last week.535installs157Pricing StrategyPricing Strategy is an agent skill packaged as a deep-dive reference on SaaS pricing models, aimed at solo and indie builders who need to charge for a product without copying a competitor blindly. It walks through how per-seat pricing works, when collaboration and headcount map cleanly to value, and when seat-based billing fights your product shape. You get concrete positioning against tools like Salesforce, Linear, Figma, and Notion, plus expansion dynamics so you understand how revenue can grow as customers hire. The material also calls out failure modes—single power-user workflows, contractors on the account, and weak seat enforcement—so you can avoid picking a model that punishes good customers. Creator/viewer and minimum-seat variants are spelled out for typical B2B adoption patterns. Use it while validating price and packaging, and again when you rethink packaging during growth.535installs158Migration ArchitectMigration Architect is an agent skill aimed at solo builders and small teams who must move production data or services without embarrassing downtime. It combines scripted tooling—migration plan generation, schema and API compatibility checking, and rollback procedure synthesis—with reference material on proven patterns, zero-downtime cutover techniques, and reconciliation strategies when two systems must temporarily disagree. Sample JSON assets illustrate database and service migrations and paired before/after schemas so agents can practice realistic diffs instead of hand-wavy “big bang” plans. The workflow assumes you already have a target architecture and need a disciplined sequence: validate compatibility, stage dual-write or shadow reads where appropriate, define measurable success checks, and keep rollback scripts ready before you flip traffic. Use it when Operate infra work forces a Postgres version jump, service extraction, or cloud relocation—not when you only need a single-table Alembic tweak with no traffic.534installs159Saas ScaffolderSaaS Scaffolder is an agent skill that surfaces SaaS architecture patterns—especially multi-tenancy models—so solo builders do not accidentally bake in the wrong isolation or cost structure. It targets indie founders and small teams planning a B2B or multi-customer product who need a structured comparison before migrations and billing integrations pile up. Use it during Validate when you are scoping MVP versus enterprise readiness, and again during Build when you are shaping backend schema and deployment boundaries. The reference explains shared-database tenancy for lowest cost and fastest launch, schema-per-tenant for moderate isolation at hundreds of tenants, and database-per-tenant for maximum isolation and tuning. Understanding those trade-offs early prevents expensive rewrites when compliance, backups, or per-tenant performance become non-negotiable.534installs160Decision Loggerdecision-logger is an agent skill built around a small Python reporter that turns markdown board-meeting decisions into actionable operational views. Solo founders and indie teams who document C-level or advisor calls in a standard decisions file can run summaries, list overdue action items, flag conflicting resolutions, filter by owner such as CMO, search by keyword like pricing, and see items due within N days. The workflow assumes decisions live under memory/board-meetings/decisions.md and that each action item carries owner, due date, review date, and completion metadata. It fits Prism’s Operate journey because shipping features without closing decision loops creates silent drift on pricing, positioning, and hiring bets. Use it after meetings in Grow or Validate when strategic choices must become dated tasks, then re-run before the next sync. The bundled script is demo-friendly for trying the parser without a populated repo.533installs161Email Sequenceemail-sequence is a reference-style agent skill that gives solo builders and micro-SaaS founders a structured playbook for lifecycle email programs without hiring a lifecycle marketer. It encodes principles such as single-purpose messages, value-led copy before hard asks, and segmentation over blast frequency. Builders use it when drafting welcome series after launch, nurture paths for trial users, onboarding checklists tied to product milestones, and short re-engagement arcs for churn risk. The skill covers recommended sequence lengths, send cadence by relationship stage, B2B versus B2C timing notes, and concrete subject-line and preview-text formulas. It fits Prism’s Grow journey because distribution alone does not retain users—email remains the default owned channel for indies shipping on tight stacks. Pair it with landing and pricing work from Validate when sequences must reflect positioning, then wire sends through your ESP of choice.533installs162Launch StrategyLaunch Strategy is an agent skill that teaches solo and indie builders how to market releases as a repeating system, not a single announcement day. It centers on the ORB framework—balancing owned channels you control, rented platforms for reach, and borrowed credibility—so every launch reinforces email lists, sites, and communities you own. The philosophy fits builders shipping SaaS, content products, or ecommerce who need distribution without enterprise marketing teams. Use it when planning a first ship, a major feature drop, or a repositioning campaign and you want channel choices, sequencing, and FOMO mechanics spelled out in plain language. It pairs naturally with landing-page and SEO skills earlier in validate and with lifecycle and analytics skills in grow once traffic lands. The skill is editorial and strategic: it does not post on your behalf but gives your coding agent a structured launch narrative aligned with how strong indie products actually get noticed.533installs163Ma PlaybookM&A Playbook is an agent skill that packages a comprehensive due-diligence checklist for mergers and acquisitions, organized by domain so you focus on items that match your deal thesis. It is aimed at solo founders and small teams contemplating buying a micro-SaaS, selling their project, or partnering on consolidation—not full investment-bank processes. The skill walks financial quality of revenue, profitability and burn, cash and liabilities, and controls, then extends into technical architecture, security, IP, contracts, HR, and go-to-market dependencies depending on what the readme exposes in your install. Use it when you have a serious LOI or term sheet conversation and need a disciplined question list for yourself, your lawyer, and your technical reviewer. It does not replace professional advisors but gives your coding agent a structured workbook to gap-fill documents, flag risks, and summarize open items before you commit capital or equity.532installs164Autoresearch AgentAutoresearch Agent is an agent skill for solo builders who want hands-off optimization instead of one-off tuning guesses. You point it at a target file or config, wire an evaluation command, and pick a metric such as p50 milliseconds, bundle size in bytes, test pass rate, or Docker build seconds. The skill documents engineering experiment domains and evaluator hooks so your coding agent can propose changes, re-run benchmarks, and chase the metric direction you set. It fits indie API and SaaS repos where search latency, frontend bundle weight, flaky CI, or slow images block shipping. Use it when you already have tests or build scripts and need a repeatable autoresearch loop rather than ad-hoc chat suggestions. It complements normal development by turning performance and reliability fixes into measurable experiments you can run overnight.531installs165Azure Cloud ArchitectAzure Cloud Architect is an agent skill that turns fuzzy "we should use Azure" conversations into requirement-backed architecture choices and Bicep-ready direction. Solo founders and small teams use it when they must pick between managed web platforms and Kubernetes, estimate monthly burn before provisioning, and align compliance labels with actual service selections. The documented workflow starts by capturing application shape, traffic, budget ceilings, team skill, availability targets, and residency needs, then drives the bundled architecture designer script to emit a recommended pattern, stacked services, tradeoffs, and a cost estimate—illustrated with App Service plus SQL and security peripherals for moderate web loads. Subsequent steps in the skill point toward template generation and pipeline setup for Azure DevOps-style delivery. On Prism, it sits in the builder journey where backend and infrastructure decisions lock in operating cost and migration risk, giving agents a repeatable script instead of generic cloud buzzwords.531installs166Email Template BuilderEmail Template Builder is a POWERFUL engineering skill for solo builders adding transactional email to a SaaS or API product. It walks agents through end-to-end communications infrastructure: choosing React Email or MJML for client compatibility, scaffolding common lifecycle messages, and integrating a provider with a single interface so you can swap vendors later. The workflow includes a dev preview server, localization with typed translation keys, optional dark mode, and deliverability hygiene so messages land in the inbox rather than spam folders. Use it when launching a new product’s email stack, migrating off a legacy sender, adding invoice or digest types, or refactoring templates for accessibility and international users. It is aimed at indie teams who want copy-paste-ready code instead of piecing together provider docs and email-client quirks by hand.531installs167Epic DesignEpic Design is an agent skill that turns a world-class scroll and motion playbook into implementation guidance for solo and indie builders who want sites to feel alive, not template-flat. It applies to landing pages, product sites, hero bands, and any layout where parallax, sticky sections, overlapping chapters, clip-path reveals, and words that illuminate on scroll should sell the story. The workflow insists on inspecting the page, judging motion hierarchy, and planning assets before coding so you are not guessing with placeholder layers. Coverage spans 45+ named techniques grouped into eight categories, all achievable without WebGL or 3D tooling—only flat assets, CSS, and JavaScript. Install it when stakeholders say immersive, premium, cinematic, or Apple-style animation, or when you need sections to bleed into each other and products to rise between bands. It pairs naturally with validate-phase landing work and launch-phase polish, but the canonical Prism shelf is Build → frontend because delivery is interface and motion code.531installs168Skill TesterSkill Tester is a thin fixture pack for validating a text-processor-style agent skill against known inputs. Indie builders who author Claude skills use it during Ship when they need deterministic sample files instead of inventing content on every run. The bundle includes multi-line prose, alphabet pangram lines, numeric and special-character rows, a small CSV with international cities, and a partial JSON statistics block that records file size and word counts. It does not implement testing logic itself—it is sample data meant to be fed into your skill under test. Pair it with your own assertion step or manual review. Confidence is moderate because the readme reads like generated fixtures rather than a full SKILL.md with invoke triggers.531installs169Tech Debt TrackerTech Debt Tracker is an agent skill packaged around a codebase scanner that turns a directory walk into an auditable debt register. Solo builders maintaining their own SaaS or API repo can invoke it to surface oversized functions, duplicated validation logic, risky hardcoded secrets, and runaway cyclomatic complexity with consistent IDs and severities. The JSON-shaped report gives a health score and debt density so you can decide what to fix before release versus what to schedule for Operate. It complements ad-hoc review chat by making items trackable (status identified, file paths, typed categories). Use it when the repo has grown past “just ship it” and you need a single pass that quantifies cleanup work for your next sprint or agent-assisted refactor session without pretending the scan replaces human judgment on product priorities.531installs170Soc2 ComplianceSOC 2 compliance is an agent skill for solo founders and tiny teams selling B2B SaaS who need a practical evidence playbook instead of hiring a compliance manager on day one. It walks through what auditors expect for Type I design reviews and Type II operating effectiveness over an observation window, including how to judge whether screenshots, logs, policies, and access reviews actually prove a control ran consistently. You get structured tables for evidence types and quality criteria so agents can help you inventory gaps, suggest automation hooks, and keep artifacts within the audit period. Canonical placement is ship security because that is when you formalize controls before enterprise deals, but the same guidance applies while operating production when you refresh quarterly evidence. It complements technical hardening skills by focusing on documentation and sampling discipline rather than Terraform or IAM specifics.530installs171Org Health DiagnosticOrg Health Diagnostic is an agent skill that runs a structured, eight-dimension organizational health assessment with traffic-light scoring and benchmark-backed drill-downs. Solo founders and indie operators wearing every hat can paste key metrics or use the bundled Python health_scorer CLI to produce a scorecard that mirrors how a full executive team would read runway, revenue, product, engineering, people, operations, security, and market health together. Use it when you are preparing for investor or advisor check-ins, feel one function is fine in isolation but the business still wobbles, or want a repeatable /health ritual instead of ad-hoc spreadsheet reviews. The skill encodes real threshold bands (for example runway green above twelve months) so agents do not invent generic advice. It is methodology-heavy process tooling with optional JSON export for integrations, aimed at small companies that lack dedicated CFO/CHRO analytics but still need board-grade clarity.529installs172AgenthubAgentHub packages dispatch prompt templates for hub-style multi-agent sessions, starting with an optimizer pattern aimed at measurable gains—latency, bundle size, memory, content quality, conversion, or research depth. Each spawned agent follows a strict iteration ritual: one focused change, run the eval command, record the metric, commit if improved otherwise revert, and publish progress to the shared board. Solo builders running Claude Code or similar agents use it when they want structured parallel experiments instead of one chat thread guessing at optimizations. The template encodes constraints such as early commits per win, switching strategy after three flat iterations, and leaving the tree in a passing-test state. It fits agent-native repos that already use `.agenthub` board paths and eval hooks.528installs173Demo VideoThe demo-video skill is a scene design system reference for solo builders who need launch-ready product films without hiring a motion studio on day one. It codifies dark-mode gradient backgrounds, semantic colors for problem versus success beats, Inter-based type ramps, and repeatable HTML scene structure at full HD. Animation pacing and easing curves keep agent-generated storyboards from feeling amateur. Voice selection ties to edge-tts personas for product, tutorial, enterprise, or consumer positioning. Use it when turning screenshots and copy into a structured reel for Product Hunt, landing pages, or onboarding—not for arbitrary long-form editing suites.528installs174Chro Advisorchro-advisor is an agent skill that encodes compensation-framework reference material for founders and operators defining how people get paid. Solo builders who stay one-person may only skim it for consultant-rate or contractor pricing context, but it shines once you hire: you get explicit choices on market percentile (P50 hybrid with equity, P75 premium talent, and related patterns), what counts as total compensation, and whether bands are visible inside the company. The skill walks through band structure, equity versus cash tradeoffs, refresh cadence, and raise processes so agents help you draft policies instead of improvised offers. It is planning and governance content, not payroll software integration. Use it when you are iterating org systems in Operate, and secondarily in Grow when compensation affects retention and lifecycle of your team. Intermediate complexity assumes you are making real hiring decisions, not learning what a salary is.527installs175Cro AdvisorCRO Advisor is a Claude skill that packages chief revenue officer frameworks for B2B SaaS builders who wear every go-to-market hat. It helps with revenue forecasting, sales model design, pricing strategy, net revenue retention, expansion revenue, churn, and sales team scaling—including when to invoke weighted pipeline models or churn analysis scripts. Triggers include designing the revenue engine, setting quotas, modeling NRR, evaluating pricing, building board forecasts, or any mention of CRO topics such as ARR growth, GRR, CAC payback, or magic number. Solo founders use it to pressure-test pricing and capacity before committing engineering scope, and again in growth phases when retention and expansion dominate. It complements product validation by making revenue assumptions explicit rather than hand-wavy spreadsheets.527installs176Marketing ContextMarketing Context is a checker skill backed by a Python script that reads your marketing context document and scores completeness from 0 to 100 using weighted sections such as product overview, target audience, problems, competitive landscape, differentiation, customer language, brand voice, and proof points. Solo builders use it before writing landing pages, lifecycle email, or LLM-generated copy so agents do not invent positioning from half-filled notes. It fits validate and scope when you are proving the idea and narrative, but the same file feeds launch distribution and SEO pages and grow content pipelines once you iterate messaging. The skill expects an existing markdown or text artifact with recognizable section markers; it does not research competitors for you—it audits what you already captured. Outcome is a quantitative score plus implicit gaps to fill, reducing rework when downstream skills or humans produce campaigns from a single source of truth.527installs177Terraform Patternsterraform-patterns is a Terraform module design reference packaged as an agent skill for solo and indie builders who own their own cloud footprint. It walks through concrete layout and HCL patterns—starting with a single-directory flat module suited to small projects and single-owner repos—so your agent does not improvise inconsistent file splits or missing backend and versions blocks. The included VPC and subnet example demonstrates how to wire required_version, provider constraints, locals for tags and prefixes, counted subnets across AZs, and validation on environment variables in one coherent package. Use it when you are standing up first infrastructure, refactoring a messy root module, or teaching an AI coding agent your house style before apply/plan cycles. It complements generic Terraform docs by focusing on module boundaries and repeatable structure that scales to later extraction into child modules, without replacing provider-specific security review or remote state runbooks.527installs178Board Deck BuilderBoard Deck Builder is an agent skill that encodes proven SaaS board-pack frameworks—especially the Point Nine style where metrics carry the story and prose only clarifies them. Solo and indie founders shipping subscription products use it when preparing quarterly board updates, Series A diligence, or advisor reviews without hiring a CFO deck consultant. The skill stresses annual thinking (ARR over MRR), Net Revenue Retention as the north-star retention metric, and operational health signals such as gross margin floors, CAC payback months, and burn multiple versus net new ARR. It includes a concrete ARR waterfall template so expansion, churn, and contraction are visible in one slide, plus benchmark bands for growth, NRR, logo churn, and runway. Because board rhythm spans reporting (Grow), fundraising proof (Validate), and ongoing governance (Operate), treat this as a multi-phase reference you pull whenever investors expect a disciplined pack—not a one-off template generator.526installs179Company OsCompany Operating System is an agent skill that walks a founder-led team through a 90-day, incremental rollout of an accountability-driven operating model inspired by EOS-style practice. It is aimed at solo builders who have grown into small teams and need one coherent system for who owns what, which metrics matter, and how leadership meets—without trying to install every ritual in the first week. The guide starts with leadership alignment and a current-state audit, assigns a single OS owner, then prioritizes Week 1–2 wins: an accountability chart workshop and a scorecard the team can actually use. Later phases extend into meeting cadence and function-level clarity across sales, marketing, product, engineering, customer success, finance, and people. Use it when informal ownership and inconsistent metrics are slowing decisions; skip it if you are a solo operator with no leadership bench to run the model.526installs180Coo AdvisorCOO Advisor is an agent skill that encodes operational cadence: which meetings exist, how async updates replace sync time, and how decisions and reporting should flow so leadership is proactive instead of reactive. Indie founders hiring their first contractors or running a five-person product team use it to copy concrete templates—daily standups, facilitation norms, and cadence principles—without hiring a full-time COO. It appears most often in operate and iterate when rituals are drifting, but the same guidance applies during build when you are forming engineering habits and during grow when lifecycle and customer rhythms need calendar space. The content is advisory prose and templates, not an automated checker or integration; you adapt formats to Slack, Loom, or your stack. Outcome is a documented rhythm that minimizes context-switching while keeping blockers visible and strategic topics on the right weekly or monthly slot.526installs181Data Quality AuditorData Quality Auditor is an agent skill that teaches procedural data-quality review grounded in missing-data theory, especially Rubin’s MCAR, MAR, and MNAR mechanisms. Solo and indie builders use it when CSVs, event streams, or warehouse tables have gaps before training models, shipping analytics, or trusting production metrics. The skill separates safe imputation paths (MCAR and conditional MAR) from MNAR cases where imputation introduces systematic bias and domain owners must decide. It emphasizes detection signals—whether null rows look like random noise or differ systematically on observed columns—and documents when to escalate rather than auto-fill. It fits builders shipping SaaS analytics, agent tools over tabular data, and API backends that ingest user-generated fields where skip patterns mirror sensitive unobserved values.526installs182Programmatic SeoProgrammatic SEO is a Launch-phase generator skill centered on a small Python tool that turns a JSON template and variable lists into every planned URL slug for template-driven organic pages—think tool-vs-competitor comparison paths at scale. Solo builders use it after they know their keyword matrix but before bulk page generation or CMS import, so agents and scripts share one canonical URL plan instead of guessing paths in chat. It supports demo mode, file-based configs, and JSON stdout for downstream content or static-site generators. Pair the output with your content templates, internal linking rules, and sitemap generation during Ship perf checks if crawl budget matters. It does not write copy or guarantee rankings; it structures URL inventory so programmatic pages stay consistent and deduplicated.526installs183Runbook Generatorrunbook-generator helps solo builders who suddenly own on-call and deploy rituals turn scattered notes into actionable runbooks. It ships template structures for deployments, incidents, and database maintenance, each with explicit verification, rollback, and escalation sections, plus a small Python generator that scaffolds a service runbook with owner, environment, and last-verified fields. A staleness template reminds you to refresh docs when Vercel, Helm, Terraform, GitHub Actions, or schema migrations change. The quarterly validation checklist pushes staging execution, rollback drills, and contact ownership updates so runbooks do not rot. Use it when preparing first production ship, after outages, or when handing ops to your future self.526installs184Spec Driven Workflowspec-driven-workflow packages acceptance-criteria patterns for solo builders who want specifications that QA and agents can execute, not prose-only roadmaps. The SKILL content centers on a repeatable Given/When/Then structure: name each criterion, state preconditions the test harness can set up, define the triggering action, and list observable outcomes without subjective language. Every acceptance criterion must reference at least one functional or non-functional requirement identifier, and multiple scenarios split into separate ACs rather than stacking contradictory When/Then blocks in one entry. The included pattern library accelerates authentication flows and similar features with concrete examples such as JWT responses, refresh tokens, and explicit HTTP status expectations. Use it while scoping a feature, drafting API contracts, or preparing test plans so downstream coding agents implement against checkable statements. It pairs naturally with implementation planning skills once the AC set is stable.526installs185Spec To RepoSpec-to-repo (Spec Parsing Guide) is an agent skill for solo builders who start from conversational or incomplete specs and need structured requirements before code exists. It instructs the agent to read the full spec once, then on a second pass fill a structured interpretation table, preferring explicit statements and strong signals over needless clarification. When the stack or database is unspecified, it applies documented defaults—Next.js and TypeScript for typical web apps, FastAPI for API-only, Flutter for mobile, Python for CLI, Go when performance is emphasized, PostgreSQL when accounts persist, SQLite for small local tools, and no database when persistence is not mentioned. That makes it ideal at Validate → scope and early Build → pm when you are deciding what to generate or scaffold. Intermediate complexity reflects judgment on inference versus asking the user. Outcomes are clearer non-negotiables, inferred features, and a stack direction you can hand to implementation skills or your agent’s repo bootstrap.526installs186Ab Test SetupA/B Test Setup is a marketing experimentation skill that walks solo builders through planning and designing A/B, split, and multivariate tests with statistical discipline. It triggers on language like A/B test, variant copy, hypothesis, conversion experiment, and statistical significance. The agent acts as an experimentation coach: clarify what you are improving, baseline conversion and traffic, tooling and timeline, then shape a test that can produce actionable winners—not open-ended tinkering. It aligns with product marketing context when a repo already has `.claude/product-marketing-context.md`. Core principles cover hypotheses, one change per test, and rigor around sample size and early stopping. Use it before you ship UI or copy changes you intend to measure, and pair with analytics-tracking when you need event or funnel implementation. Intermediate complexity reflects traffic math and tooling choices without requiring a dedicated data science team.525installs187Board Meetingboard-meeting is a facilitation skill for solo founders and tiny teams who simulate a real board or executive staff meeting inside an agent workspace. Instead of one model playing every function, the guide enforces a six-phase protocol with disciplined Phase 2 contributions: maximum five material points per role, mandatory recommendations, confidence scores, and falsifiable 'what would change my mind' lines so CFO or legal personas cannot bury the agenda in thousand-word essays. Chief of Staff instructions explain how to trim overrun while preserving recommendations in a raw log, and Phase 3 guidance clarifies that the Executive Mentor exists to surface tradeoffs, not to rubber-stamp harmony. Reference it when you need structured debate on roadmap bets, budget, hiring, or go-to-market before you commit engineering weeks. It complements planning skills that produce specs, but this one optimizes how diverse agent roles argue and converge.525installs188Cpo Advisorcpo-advisor packages a CPO-style PMF Playbook for solo and indie builders who need to prove demand before pouring months into code. It walks you from defining real product-market fit—organic discovery, emotional reliance, and referral behavior—through finding your highest-PMF segment by contrasting retained versus churned cohorts and pressure-testing with short customer calls. The skill then drives measurement across multiple complementary signals so one metric cannot mislead you, and it warns against common false positives such as contract lock-in or praise without retention. Use it when you have early users or a prototype and must decide whether to narrow positioning, pivot, or scale distribution. It pairs naturally with validation landing tests and later grow-phase analytics, but the canonical home is validate/scope because fit is the gate before full product investment.525installs189Focused Fixfocused-fix is an agent skill for deep-dive feature repair when an entire module or user-visible capability is unreliable, not a lone stack trace. The workflow expects systematic exploration: map dependencies, follow logs, run and interpret tests, and align fixes across every file that participates in the feature. Solo builders invoke it when prompts like “make checkout work” or “the billing module is broken” mean days of partial patches failed. The skill explicitly defers narrow symptom chasing to systematic-debugging; use focused-fix when the failure surface is wide and integration-shaped. It fits backend-heavy SaaS, agent tooling, and API features where one broken contract ripples through UI, jobs, and tests.525installs190Helm Chart BuilderHelm Chart Builder is a reference-oriented agent skill for solo and indie builders who ship backends and APIs to Kubernetes without a platform team. It encodes how to lay out Chart.yaml, values, .helmignore, and templates so releases stay reproducible and upgrade-safe. You use it when you need a minimal production chart or a fuller stack with ingress, autoscaling, pod disruption budgets, network policies, and JSON Schema–backed values. The skill emphasizes copy-paste _helpers.tpl conventions, consistent naming overrides, and test hooks—reducing chart drift across services. It fits Claude Code, Cursor, and similar agents during infra hardening or when converting a docker-compose prototype into something your cluster operator can install with helm upgrade --install.525installs191Paid Adspaid-ads is a reference-style agent skill packed with ad copy templates for high-converting primary text. Solo builders use it when they are ready to spend on distribution but lack a copywriter: the skill walks through Problem-Agitate-Solve, Before-After-Bridge, social-proof-led openers, feature-to-benefit bridges, and direct-response layouts with worked examples (reporting automation, approvals, collaboration). It does not replace channel setup, pixel configuration, or budget strategy—it gives repeatable sentence patterns your agent can adapt to your product name, offer, and compliance constraints. Best invoked when you have a clear ICP and offer and need several variant drafts for A/B tests or platform character limits.525installs192Apple Hig Expertapple-hig-expert is a reference skill for solo developers shipping on Apple platforms who need accessibility and HIG compliance baked into implementation conversations. It organizes work around the four pillars—perceivable, operable, understandable, and robust—with actionable checks such as descriptive VoiceOver labels, dual encoding beyond color alone, minimum hit areas, and standard navigation patterns users already recognize. Technical notes emphasize Dynamic Type–safe stacks and contrast minimums so agents do not ship pretty screens that fail review or real assistive-tech use. It is aimed at indie app builders who cannot afford a dedicated accessibility QA pass on every sprint but still want App Store–credible quality. Invoke while designing or refactoring screens, not after launch when retrofitting labels and layout is expensive.524installs193Competitor Alternativescompetitor-alternatives teaches solo founders and marketers how to architect competitor data once and reuse it across many “vs” and alternative URLs. Instead of rewriting Notion or Monday comparisons on every draft, you maintain one markdown or YAML profile per competitor with pricing, feature scores, positioning, and fair weakness callouts. That structure feeds SEO comparison pages, sales battlecards, and validate-phase positioning workshops without contradictions between posts. The skill emphasizes intellectual honesty—document real strengths alongside gaps your product solves—so AI-generated copy does not read like strawman marketing. It pairs naturally with idea-phase competitive research and later launch-phase SEO pages, but it does not replace live pricing API sync or automated SERP tracking. Intermediate complexity: you need a content repo and willingness to maintain profiles as competitors ship changes.524installs194Full Page ScreenshotFull Page Screenshot is a Node.js CLI skill built around full-page-screenshot.mjs that drives Chrome through the Chrome DevTools Protocol to capture entire pages—not just the viewport. Solo builders use it when they need trustworthy visuals for ship checks, changelog docs, or launch posts without clicking through manual capture tools. Run --check to validate Node 22+ and debugging port, --list to pick an open tab, or --url to open a URL, wait for load, capture, and close. You can also target an existing tab by ID for iterative frontend work. Options tune viewport width, device pixel ratio, and load timeout so retina marketing shots and wide dashboard layouts stay consistent. It fits agent workflows that already shell out to local tools: the agent supplies URLs and paths, you keep Chrome debugging on locally. Not a cloud browser farm—runs on your machine with real Chrome.524installs195Git Worktree ManagerGit Worktree Manager is a production-oriented skill for solo builders and tiny teams who run more than one feature branch at a time without port collisions or drifted environment files. It bundles two Python CLIs: one to create and prepare named worktrees from a base branch with optional installs, and one to audit stale or merged trees before removal. Docker Compose override guidance keeps each tree runnable on its own port. The skill targets agent-assisted DevOps on your laptop or CI-adjacent scripts, not hosted Git hosting features. Use it when context-switching hurts velocity but you still want isolated test servers per branch.524installs196Intl Expansionintl-expansion is a strategy skill for solo founders and indie operators weighing whether and how to sell into new countries. It packages a practical sequence—from scoring addressable markets and regulatory burden through choosing entry modes, localization plans, regional GTM, and team shape—so you do not treat international growth as a vague “translate the app” task. The scoring matrix makes tradeoffs explicit: size, incumbents, compliance cost, cultural distance, existing inbound traction, and operational complexity. Use it when researching expansion options, planning localization investments, or aligning compliance timelines with launch. It sits early in the journey for discovery and pairs with validate-phase scoping when you narrow to one region. Written for agent-assisted planning with MIT-licensed metadata version 1.0.0 in the upstream skill pack.524installs197Marketing IdeasMarketing Ideas is an agent skill that surfaces a structured playbook of 139 marketing approaches organized by category, with a deep Content & SEO section and adjacent growth patterns like repurposing and data-driven content. Solo and indie builders install it when they need credible tactic ideas for launches, early traction, or scaling distribution without hiring a full growth team first. The skill works as editorial reference material: your agent can map your niche, constraints, and current funnel stage to specific numbered ideas—low-competition keyword plays, template-driven programmatic pages, glossaries for “what is X” queries, and refresh cycles for existing posts. It fits the Launch and Grow phases of the Prism journey because most entries assume you are shipping something real and now need repeatable ways to be discovered. Complexity stays beginner-friendly because the value is selection and prioritization, not executing every tactic at once. Pair it with site-architecture or SEO audit workflows when you move from brainstorming to implementation.524installs198Release Managerrelease-manager is an agent skill aimed at solo and indie builders who ship frequently and need disciplined release hygiene without a dedicated release engineer. It helps you interpret recent commit history—features, fixes, and documentation—and shape that into coherent release planning: what ships, how to describe it, and what operational steps precede go-live. The skill aligns with conventional commit narratives and issue references so notes stay traceable for users and future you. Use it when a merge window closes and you must decide version semantics, draft notes, and coordinate launch tasks. It complements testing and security skills rather than replacing them; you still own verification before tag. Prism lists it for builders on Claude-family agents who want repeatable release rituals from repository facts.524installs199Roadmap CommunicatorRoadmap Communicator is an agent skill for solo builders and small teams who need to communicate direction without drowning stakeholders in backlog detail. It guides you through choosing among Now/Next/Later, fixed timeline, and theme-based roadmaps depending on uncertainty and commitment level, then tailoring the same underlying plan into executive outcome summaries, engineering scope and dependency updates, and customer-facing value narratives. The skill also covers user-facing and internal release notes, changelog extraction from git history, and structured feature announcements so launches feel coordinated rather than improvised. Use it whenever you are preparing roadmap presentations, board updates, sprint-visible sequencing for engineers, or timing windows for customers. It does not replace prioritization frameworks or issue tracking; it makes what you already decided legible to each audience with consistent tone and trade-off transparency.524installs200Site ArchitectureSite Architecture is an agent skill packaged as an Internal Linking Playbook: it teaches patterns for crawlability, equity distribution, and topical authority through deliberate link graphs. Solo builders and small teams use it when a content site, docs hub, or SaaS marketing site has pages that rank poorly because nothing important links to them, or because Google cannot reach deep URLs efficiently. The playbook contrasts architectures—hub-and-spoke clusters for blogs and solution pages, linear chains for tutorials and documentation—and spells out who should link to whom with sensible anchor-text guidance. It is intermediate in complexity because you must map your real URL inventory and pillar keywords, not just paste a diagram. In the Prism journey it appears at Launch for SEO strategy and again in Build when you wire routes, nav, and in-body links in your app or CMS. Agents apply it to produce a linking plan you can implement in Next.js, static site generators, or headless CMS templates without guessing structure ad hoc.524installs201Executive MentorExecutive Mentor is an agent skill centered on Board Dynamics—how founders who report to a board manage directors, expectations, and the slow erosion of trust that precedes leadership changes. It is aimed at solo founders who have graduated into CEO roles with formal governance: operator directors who want operational candor, and financial investor directors who watch growth, burn, and fund-return math. The skill is advanced because mistakes are high stakes and context-dependent; it does not replace lawyers or coaches but gives structured mental models for prep, communication style, and reading disengagement. In Prism’s journey it sits in Operate when you are iterating the business with external oversight, and it also touches Grow when scaling changes what the board measures. Indie hackers without any board can skip it; builders preparing a seed or Series A round may use it in Validate or Launch to anticipate director types before seats are filled. Use it before quarterly board decks, difficult metric conversations, or when you sense a director going quiet.523installs202Scenario War Roomscenario-war-room gives solo SaaS founders a structured reference for thinking in futures rather than single-point forecasts. It adapts Shell’s scenario planning—mutually exclusive storylines, a two-by-two on critical uncertainties, memorable scenario names, predetermined elements, and early indicators—into an early-stage raise-and-runway frame such as Blue Ocean, Ramp Carefully, Scrappy Growth, and Survival Mode. The Monte Carlo section is deliberately lightweight: identify a handful of variables, assign plausible ranges, and reason about distribution outcomes without standing up simulation tooling. Use it when fundraising timing, market velocity, or burn forces tradeoffs that ad-hoc optimism hides. It complements validation work where you need shared language for “what we do if X” before locking roadmap or pricing.523installs203Signup Flow CroSignup-flow-cro gives solo builders a practical conversion playbook for registration flows on SaaS landing pages, waitlists, and lightweight ecommerce accounts. It centers on four principles: collect only what you truly need before first value, show product benefit before asking for commitment, reduce perceived effort with progress and defaults, and remove uncertainty about time and next steps. The reference goes field by field—email, password, name—with concrete UX fixes such as inline validation, typo correction for common domains, strength meters instead of punitive rules, and passwordless alternatives where appropriate. It is aimed at founders tuning MVP funnels rather than enterprise IAM compliance projects. Pair it with analytics after changes ship so you can measure lift on completion rate and time-to-first-session rather than guessing from copy tweaks alone.523installs204Agent ProtocolAgent Protocol is an agent skill that codifies inter-agent messaging patterns for solo builders and small teams running multiple specialist agents—sales, finance, marketing, or custom roles—in one workflow. Instead of unstructured handoffs, it shows how to open a planning thread, invoke peers with scoped questions, capture responses with key findings, supporting data, confidence, and caveats, then chain follow-ups. The included Revenue Planning Chain demonstrates realistic numbers-driven collaboration for ARR targets, runway, pipeline coverage, and CAC. Use it while designing Claude Code or Cursor multi-agent setups, custom orchestration layers, or internal playbooks where agents must cite constraints and pass context forward. It is methodology for reliable multi-step reasoning, not a hosted protocol server.522installs205Form Croform-cro is a conversion-focused reference skill for solo builders optimizing user-facing forms on landing pages, pricing pages, and early funnels. It encodes practical CRO rules: every field has a measurable completion cost, perceived value must beat effort, and cognitive load should stay low through grouping, defaults, and clear labels. The guide walks field-by-field tactics—single email fields without confirmation, when to use one name field versus split names, optional phone with explained rationale, post-submit enrichment for company data, and dropdown patterns that do not trap mobile users. It fits indie SaaS, content products, and ecommerce flows where a single form is the gate to trials, demos, or newsletters. Use it while you are still shaping validation pages or before scaling paid traffic, so you are not paying to send users into high-friction capture flows. The skill is reference-oriented rather than a generator: your agent applies the checklist to your specific form copy and field list.522installs206Onboarding CroOnboarding-cro packages a Python activation funnel analyzer for solo SaaS founders who need hard numbers on where users stall before they rewrite UX copy blindly. You supply a JSON list of named steps with user counts—signup, verification, profile setup, first action, aha moment, activated—and the tool computes conversion from the top of funnel, step-to-step drop rates, and which step loses the most people. It runs in demo mode without data or from funnel.json with optional JSON stdout for agents. Use it in Grow when you are sizing onboarding experiments, arguing for one CRO bet over another, or feeding lifecycle retros. It complements qualitative session review by ranking leaks by magnitude.522installs207Prompt Engineer ToolkitPrompt Engineer Toolkit is an agent skill for solo and indie builders who rely on LLMs for marketing output but cannot trust vibe-based prompting after every model or instruction tweak. It walks you from ad-hoc drafts to production assets: structured test cases, quantitative adherence and safety scoring, immutable version history with behavior-impacting diffs, and reusable templates tuned for ad copy, email, and social. Use it when launching a new LLM-powered marketing feature, when quality slips after a model change, when multiple people edit the same prompts, or when you need a defensible pick for which prompt ships. The emphasis is measurable quality and regression safety over intuition, which matters when one person owns product, growth, and content. It complements generic chat rewriting by adding governance patterns teams usually bolt on later.522installs208Self EvalSelf-eval is a Claude Code skill that fixes unreliable AI self-grades by separating how hard the task was from how well it was executed, then combining them through a matrix the model cannot rationalize away. After you finish a task, review, or work block, it forces devil’s advocate reasoning in both directions, resolves the tension, and logs results for longitudinal calibration. Solo builders benefit when every session otherwise ends with a polite four-out-of-five, when you want honest closure before context resets, or when you need to compare effort and quality across weeks of shipping alone. Persistence to `.self-eval-scores.jsonl` powers anti-inflation warnings. It does not replace peer review or automated tests—it makes your agent’s self-narrative measurable so you can decide whether to rework, ship, or escalate.522installs209Churn Preventionchurn-prevention packages a small Python churn impact calculator for solo SaaS founders who need a credible revenue story before changing cancellation flows, save offers, or dunning. You supply monthly recurring revenue, overall churn percentage, the voluntary share of churn, current and target save rates for voluntary exits, and current and target recovery rates for involuntary churn, plus average customer MRR. The script computes how much MRR leaves each month, how much you already save or recover, and how much additional MRR you could retain if you hit target rates—giving a quantitative basis for roadmap decisions without a full spreadsheet build. It fits the Grow phase when retention experiments need a shared assumption set across product, support, and billing changes. The skill is code-forward rather than a narrative playbook; pair it with your real cohort data and checkout analytics when you move from model to implementation.521installs210Monorepo NavigatorMonorepo-navigator is a reference skill for solo and small teams who outgrew a single package but do not want tribal knowledge buried in Slack threads. It catalogs proven layouts—classic apps and packages trees, domain-bounded folders, and service monorepos with shared libs—and spells out dependency rules that keep deployables from importing each other casually. On the automation side it connects those structures to Turborepo pipeline patterns in `turbo.json`, including topological `dependsOn`, artifact outputs, environment propagation, and splitting lint, typecheck, and test tasks so CI failures localize quickly. Release guidance covers Changesets-style versioning, automated publishes, and prerelease channels for unstable shared libraries. Invoke it when you are standing up or refactoring a monorepo during Build, when you tighten Ship CI to affected projects only, or when Operate demands clearer ownership between services and shared packages. It is meta architecture guidance, not a code generator; your agent applies the patterns to your stack.521installs211Schema MarkupSchema Markup is an agent skill that hands solo builders production-ready JSON-LD snippets for frequent content and utility page types. Instead of guessing required fields for BlogPosting, HowTo, and related schema.org shapes, you paste patterns, swap ALL_CAPS placeholders for real headlines, URLs, and image dimensions, then validate before deploy. It fits indie sites, blogs, and SaaS marketing pages where you cannot afford rich-result regressions or invalid structured data. Use it when shipping or refreshing landing pages, blog posts, tutorials, or FAQ-style guides where search visibility and snippet eligibility matter. The skill emphasizes testing at rich-results.google.com so you catch missing publisher logos, date fields, or image aspect issues early—cheap insurance compared to debugging indexation after launch.521installs212Team CommunicationsTeam communications is an agent skill for writing 3P updates—Progress, Plans, and Problems—aimed at executives, leadership, and teammates who need a fast weekly pulse. Solo and indie builders still benefit when they report to investors, a part-time cofounder, contractors, or a small distributed crew: the skill keeps updates short, scannable, and honest about blockers instead of burying them in chat threads. It asks you to confirm the team name, then balances what shipped against what is next and what is slowing you down, with guidance that larger scopes use meatier bullets while small teams stay concrete. You can invoke it during active build cycles, before investor check-ins, or when operating a live product and need a consistent external narrative. It is a template-driven writing workflow, not a analytics dashboard—pair it with your issue tracker or git history when tools are available.520installs213Context EngineCompany Context Engine is the shared memory layer for Alireza Rezvani’s C-suite advisor skill family. At session start it checks for ~/.claude/company-context.md, evaluates the Last updated field against a 90-day staleness rule, and either loads fresh context or warns you before proceeding stale. It parses stage, goals, and operating facts into working memory so CFO, CMO, and other advisors answer with your company in mind—not generic playbooks. The skill also enriches context during conversations and applies anonymization before data leaves the agent to external APIs. Solo founders and indie operators who run multi-advisor workflows install it so every executive-style skill starts from the same trusted profile. Use it whenever you open a C-suite session; refresh when the file is missing or past the freshness window.519installs214Internal NarrativeInternal-narrative is an agent skill that equips solo founders and small teams with reusable storytelling frameworks for business communication. It documents Barbara Minto-influenced SCR—Situation, Complication, Resolution—for updates where audiences already share context, illustrated with a concrete investor quarter narrative that names missed ARR, delayed deals, and the corrective pipeline methodology. It also outlines Problem-Solution-Evidence for pitches and launches: describe the world as it is, what is broken, then proof that your approach works. The skill is procedural editorial knowledge, not a copy-paste marketing blast generator; it optimizes for credibility under scrutiny, which matters when one builder wears product, sales, and fundraising hats. Use it whenever you must align internal teams and external stakeholders on the same through-line—from validation conversations through launch announcements and ongoing growth reporting.519installs215Culture ArchitectCulture Architect packages a culture playbook reference built around the Netflix Culture Deck and related frameworks for measuring and evolving how a company actually behaves. Solo founders who are about to make their first hires—or indie teams hitting five to fifteen people—can use it to stress-test whether slogans like freedom and responsibility are backed by hiring bar, performance context, and manager habits. The skill is editorial and conceptual rather than a code integration: it contrasts mechanisms that work (adequate performance policies stated clearly, publicly descriptive values) with patterns that often fail when copied blindly (unlimited vacation without modeling, keeper test without coaching). It fits Prism builders who treat culture as an operational system, not HR theater, while acknowledging that framing must adapt outside Silicon Valley defaults.518installs216Free Tool StrategyFree Tool Strategy is an agent skill that packages a launch playbook for indie builders who ship free web tools and need distribution to compound. It walks through pre-launch SEO foundations (keyword, URL slug, schema, internal links), a quality gate so the tool actually converts (few inputs, mobile, shareable results, gated capture after value), and a multi-week seeding mindset across channels rather than a single LinkedIn post. Solo founders use it when a calculator, generator, or checker is live but traffic flatlines after the first spike. The skill aligns tool pages with search intent and backlink-friendly positioning so Launch-phase work connects to long-term Grow traffic without inventing install counts or audit outcomes.518installs217Paywall Upgrade CroPaywall-upgrade-cro is an agent skill for solo and indie builders who need conversion-focused guidance on in-app paywalls, upgrade screens, upsell modals, and feature gates. It applies when users mention paywalls, upgrade modals, freemium conversion, trial expiration screens, or limit-reached prompts—distinct from optimizing standalone marketing pricing pages. The workflow starts with an initial assessment: optionally load product marketing context from your repo, then clarify upgrade context, product model (free vs paid, triggers, baseline conversion), and the moment in the user journey when the screen appears. You get structured CRO-oriented recommendations aligned to moments where value is already proven, which is critical for SaaS and mobile products that depend on self-serve upgrades rather than sales calls. Use it while iterating lifecycle monetization before major pricing experiments or redesigns.518installs218Popup Cropopup-cro is a reference skill for solo and indie builders who run marketing sites, landing pages, or content funnels and need popups that convert instead of annoy. It frames conversion work around three non-negotiables: show offers when intent is real, make the benefit obvious in one glance, and never trap users—easy dismiss and remembered preferences matter as much as the headline. The reference walks six trigger families from crude time delays through scroll depth, exit intent, user-initiated clicks, multi-page journeys, and high-intent behaviors like cart or pricing revisits, with practical defaults such as 25–50% scroll and 30–60 second engagement windows rather than instant interruption. Use it while planning or refactoring modals on validate landing experiments, launch distribution pages, or grow lifecycle capture flows so agents and humans share the same CRO vocabulary before touching HTML, scripts, or A/B tools.518installs219Referral Programreferral-program is an agent skill that packages a measurement framework for SaaS referral loops: which numbers to track weekly, how to calculate them, and what “good” looks by stage. Solo founders use it when a referral feature exists but growth feels opaque—low participation, weak conversion, or rewards nobody redeems. The skill emphasizes a small stack of actionable metrics from program awareness through K-factor, with benchmarks for early ARR bands so you do not over-optimize vanity counts. It complements product work rather than replacing it; you still need landing pages, in-app prompts, and payout rails elsewhere. Invoke it when planning dashboards, reviewing growth experiments, or writing specs for referral analytics instrumentation.516installs220Strategic AlignmentStrategic Alignment is an agent skill playbook for solo and indie builders who wear the founder and PM hat and need discipline borrowed from scaled teams without hiring a strategy office. It teaches how to compress strategy onto one page so it is clear enough to cascade, then run structured workshops where leaders translate company OKRs into department contributions and surface gaps before anyone writes code. The one-page filter forces a six-word vision, three owned priorities, an honest list of what you are not doing, and a small metric set—because without explicit deprioritization every squad invents its own roadmap. Follow-on sections in the skill address drift detection and maintaining alignment over time, which matters when you ship fast and context slips between async updates. Use it when you are planning a quarter, reconciling conflicting initiatives, or feeling “everyone is busy but nothing strategic moves.” It is methodology and templates, not an integration; pair it with your existing issue tracker and OKR tooling.516installs221Command Guidecommand-guide is a Claude Code meta skill that acts as a selection layer over slash commands, bundled agents, and related skills so you do not burn context experimenting with the wrong entry point. It targets solo builders who live in Claude Code when implementing features, fixing bugs, reviewing diffs, shrinking context, or running loops—but feel unsure whether /plan, /tdd, /compact, or a specialist agent is appropriate. The documented quick decision flowchart branches on request type: new features toward /plan and planner, bugs toward /tdd or build-error-resolver, reviews toward code-reviewer, long contexts toward /compact, documentation toward /docs, and security toward security-reviewer. Because it is pure routing knowledge, it stays useful across Validate planning chats, Build implementation, Ship review and testing, and Operate firefighting whenever you ask which command or agent fits. It does not execute tasks itself; it recommends the next Claude Code surface so your session stays intentional.513installs222Tc TrackerTC Tracker is an agent-oriented change-management workflow backed by small Python scripts that initialize a docs/TC workspace, mint sequential technical-change records, and enforce updates through a state machine with append-only history. Solo builders shipping with Claude Code or Cursor use it when multiple sessions touch the same feature branch and chat context alone is not enough to remember what was done, what files changed, and what blocked the next step. The handoff fields (progress, next actions, blockers) are designed so a fresh session can reload in_progress work without re-reading the entire repo. It pairs naturally with implementation and review phases: you open a TC when you start a feature, log file deltas as the agent edits, and write a handoff before stopping. It is not a ticketing SaaS; it is filesystem-native discipline for indie teams of one.513installs223Code Tourcode-tour teaches your agent to author Microsoft CodeTour-compatible `.tour` files: scripted, persona-specific stories that jump to real files and line numbers in VS Code. Solo maintainers and tiny teams use it when onboarding a future you, explaining how a subsystem works, or packaging a PR or incident review as a replayable walkthrough. The skill enforces discovery first—README, config, folder map, and verified paths—so tours do not point at phantom modules. It is ideal when documentation must feel like a guided tour rather than a static wiki page. Because the same pattern serves PR review during Ship and root-cause tours during Operate, treat Build docs as the home shelf while reusing the skill whenever someone needs a structured code narrative.512installs224Statistical AnalystStatistical Analyst is an agent skill that equips coding agents with rigorous frequentist testing vocabulary for solo builders running A/B tests, funnel changes, and pricing experiments. The bundled reference explains null and alternative hypotheses, p-values versus effect certainty, Type I and Type II errors, and concrete tests such as the two-proportion z-test for binary outcomes with sample-size assumptions. Use it when you have control and treatment data and need help picking a test, stating assumptions, or explaining results without treating p=0.03 as "97% chance the effect is real." It supports SaaS and API products where conversion metrics drive decisions. Complexity is advanced because statistical misuse is costly; the skill is procedural knowledge for analysis and review, not a replacement for pre-registered experimentation platforms or legal/compliance review of regulated trials.512installs225Ai SecurityAI Security is an agent skill for solo builders shipping LLM features, copilots, or tool-using agents who need a structured lens on adversarial risk instead of ad-hoc prompt hygiene checklists. It aligns controls and detection ideas to MITRE ATLAS—the AI counterpart to ATT&CK—covering prompt injection, indirect injection via retrieved content, jailbreak framing, tool invocation abuse, data exfiltration, and training-data poisoning scenarios. Use it while hardening before launch in Ship and again in Operate when you review logs or fine-tuning pipelines. The skill emphasizes regex and pattern-based detection hooks plus risk scoring for inference abuse, so you can document what you monitor and why. It does not replace a full red team or compliance program, but it gives indie teams a credible vocabulary and starter coverage map when investors or customers ask how you handle AI-specific threats.511installs226Meeting AnalyzerMeeting-analyzer is a journey-wide agent skill that ingests meeting transcripts and recordings metadata, inventories files in a directory, parses common speaker-label formats (VTT/SRT cues, bracketed names, diarized lines), and produces concrete, quoted evidence for communication patterns, leadership behaviors, and interpersonal dynamics. Solo founders and indie builders use it when they want coaching without hiring a communications consultant—after investor calls, customer discovery, standups, or conflict-heavy syncs. The workflow starts with ingest and inventory, resolves which speaker is “you” when labels are ambiguous, then analyzes ratios, fillers, avoidance, and recurring anti-patterns before actionable feedback. It is intentionally triggered by informal phrasing like “how do I come across in meetings” or uploads from Granola, Otter, Fireflies, or Zoom. Output is behavioral insight backed by transcript citations, suitable for personal retrospectives or lightweight team retros when you own the recordings.495installs227Grill MeGrill-me is an agent skill package for solo and indie builders who want plans interrogated instead of rubber-stamped. Companion stdlib Python scripts scan plan documents for decision branches, generate forcing questions with recommended answers and dependency-aware ordering, and persist multi-turn grill state under ~/.grill_sessions/ so you can resume across days or parallel projects. The cs-grill-master persona enforces a relentless, codebase-exploration-first voice and refuses bundled questions, which keeps ambiguity visible until you explicitly lock each decision. Use it whenever you have a spec, RFC, or implementation outline and need structured Socratic pressure before committing engineering time. It pairs with slash-command invocation for Claude Code-style workflows and produces auditable decision trails suitable for handoff to implementation or review skills.474installs