am-will/codex-skills
20 skills18.9k installs20.2k starsGitHub
Install
npx skills add https://github.com/am-will/codex-skillsSkills in this repo
1Frontend Designfrontend-design is a skill from am-will/codex-skills that guides coding agents to create distinctive, production-grade frontend interfaces. The skill activates when developers ask to build web components, pages, or full applications and emphasizes real working code with exceptional attention to aesthetic detail and creative choices. It explicitly targets avoidance of generic AI slop aesthetics—uniform gradients, predictable layouts, and templated component patterns. Developers provide requirements covering purpose, audience, and technical constraints; the skill returns polished HTML, CSS, and component code tuned for visual quality. It suits teams who want agent-generated UI that reads as intentionally designed rather than auto-generated.1.4kinstalls2Frontend Responsive Design StandardsThe planner skill is designed for create phased implementation plans with research, clarification, sprints, and atomic tasks. Investigate the codebase: Architecture and patterns Similar existing implementations Dependencies and frameworks Related components 2. Analyze the request: Core requirements Challenges & edge cases Security/performance/UX considerations Phase 1: Clarify Requirements Before doing ANY documentation search: clarify requirements with user. Invoke when the user says make a plan, break down tasks, /planner, or outline implementation steps.1.2kinstalls3PlannerThe planner skill is designed for create comprehensive, phased implementation plans with sprints and atomic tasks. Use when user says: "make a plan", "create a plan", "plan this out", "plan the implementation",. Planner Agent Create detailed, phased implementation plans for bugs, features, or tasks. Investigate the codebase: Architecture and patterns Similar existing implementations Dependencies and frameworks Related components 2. Invoke when the user user says: "make a plan", "create a plan", "plan this out", "plan the implementation", "help me plan", "design a plan", "draft a plan", "write a plan", "outline the steps", "break this down into tasks", "what's the plan for", or any similar planning request.1.2kinstalls4Read GithubThe read-github skill is designed for read and search GitHub repository documentation via gitmcp.io MCP service. Read GitHub Docs Access GitHub repository documentation and code via the gitmcp.io MCP service. fetch_{repo}_documentation - Fetch entire documentation. Invoke when the user asks about read github or related SKILL.md workflows.1.2kinstalls5Parallel TaskThe parallel-task skill is designed for only to be triggered by explicit /parallel-task commands. Parallel Task Executor You are an Orchestrator for subagents. Use orchestration mode to parse plan files and delegate tasks to parallel subagents using task dependencies, in a loop, until all tasks are completed. Invoke when the user asks about parallel task or related SKILL.md workflows.1.2kinstalls6Llm CouncilThe llm-council skill is designed for orchestrate a configurable, multi-member CLI planning council (Codex, Claude Code, Gemini, OpenCode, or custom) to produce independent implementation plans, anonymize and. LLM Council Skill Quick start Always check for an existing agents config file first ($XDG_CONFIG_HOME/llm-council/agents.json or ~/.config/llm-council/agents.json). If none exists, tell the user to run ./setup.sh to configure or update agents. Invoke when the user you need a robust, bias-resistant planning workflow, structured JSON outputs, retries, and failure handling across multiple CLI agents.1.2kinstalls7Plan HarderThe plan-harder skill is designed for use when user specfically says 'plan harder'. Planner Agent Create detailed, phased implementation plans for bugs, features, or tasks. You make phased implementation plans with sprints and atomic tasks. Invoke when the user user specfically says 'plan harder'.1.2kinstalls8Agent BrowserThe agent-browser skill is designed for a fast Rust-based headless browser automation CLI with Node.js fallback that enables AI agents to navigate, click, type, and snapshot pages via structured commands. Agent Browser A fast Rust-based headless browser automation CLI with Node.js fallback that enables AI agents to navigate, click, type, and snapshot pages via structured commands. Wait and Timing Advanced Control Sessions Run multiple isolated browser instances. Invoke when the user asks about agent browser or related SKILL.md workflows.1.2kinstalls9Markdown UrlThe markdown-url skill is designed for route any website you need to visit through markdown.new by prefixing the URL. markdown.new URL Prefix Whenever you need to visit a website URL, you must: 1) rewrite it to go through https://markdown.new/, then 2) visit the rewritten URL, and 3) pull the Markdown content from that page for use in your answer/notes. Normalize the destination into an absolute URL with a scheme (https:// preferred). Invoke when the user asks about markdown url or related SKILL.md workflows.1.2kinstalls10Gemini Computer UseThe gemini-computer-use skill is designed for build and run Gemini 2.5 Computer Use browser-control agents with Playwright. Use when a user wants to automate web browser tasks via the Gemini Computer Use model, needs an. Create a virtual environment and install dependencies: 3. Run the agent script with a prompt: Browser selection Default: Playwright's bundled Chromium (no env vars required). Invoke when the user a user wants to automate web browser tasks via the Gemini Computer Use model, needs an agent loop (screenshot → function_call → action → function_response), or asks to integrate safety confirmation for risky UI actions.1.2kinstalls11Openai Docs SkillThe openai-docs-skill skill is designed for query the OpenAI developer documentation via the OpenAI Docs MCP server using CLI (curl/jq). Use whenever a task involves the OpenAI API (Responses, Chat Completions, Realtime,. OpenAI Docs MCP Skill Overview Use the OpenAI developer documentation MCP server from the shell to search and fetch authoritative docs. Always do this for OpenAI platform work instead of relying on memory or non-official sources. Invoke when the user asks about openai docs skill or related SKILL.md workflows.1.2kinstalls12Swarm PlannerThe swarm-planner skill is designed for [EXPLICIT INVOCATION ONLY] Creates dependency-aware implementation plans optimized for parallel multi-agent execution. Swarm-Ready Planner Create implementation plans with explicit task dependencies optimized for parallel agent execution. Explore Codebase: Investigate architecture, patterns, existing implementations, dependencies, and frameworks in use. Invoke when the user asks about swarm planner or related SKILL.md workflows.1.2kinstalls13Vercel React Best PracticesThe vercel-react-best-practices skill is designed for react and Next.js performance optimization guidelines from Vercel Engineering. This skill should be used when writing, reviewing, or refactoring React/Next.js code to ensure. Vercel React Best Practices Comprehensive performance optimization guide for React and Next.js applications, maintained by Vercel. Contains 45 rules across 8 categories, prioritized by impact to guide automated refactoring and code generation. Invoke when the user asks about vercel react best practices or related SKILL.md workflows.1.2kinstalls14Role CreatorRole Creator enables developers to define and update custom Codex agents through standalone TOML files stored globally (~/.codex/agents/<name>.toml) or per-project (<project>/.codex/agents/<name>.toml). The skill enforces a strict input-collection workflow: confirm name, description, developer_instructions, model, reasoning_effort, scope, and output path before writing. Generated files contain required keys (name, description, developer_instructions) plus optional fields (nickname_candidates, model, sandbox_mode, web_search, mcp_servers). Validation occurs post-write; ready-to-run JSON payloads are returned for immediate agent spawning.1.2kinstalls15Parallel Task SparkParallel Task Executor (Sparky) orchestrates distributed task execution by parsing plan files, building dependency graphs, and launching Sparky subagents in parallel waves. It manages task blocking/unblocking based on dependencies, enforces RED-phase testing before implementation, validates work through test evidence or documented alternatives, commits changes atomically per task, and updates plan logs. Designed for coordinated multi-agent development workflows where tasks have complex interdependencies.1.1kinstalls16Super Swarm SparkThe super-swarm-spark skill orchestrates parallel Sparky subagents parsing markdown plan files into delegated tasks with a rolling scheduler keeping up to twelve agents active. Triggered only by explicit super-swarm-spark commands. Step one extracts plan file path and optional task ID subset. Step two parses task subsections like T1 headers capturing IDs, linkage metadata, descriptions, locations, acceptance criteria, and validation steps. Step three builds per-task context packs with canonical file paths, exact new filenames, neighboring task conflicts, naming constraints, and cross-task expectations; subagents must not invent alternate filenames. Step four launches sparky agent_type tasks immediately up to twelve concurrent slots, validating each completion, updating the plan file, and filling open slots until pending and running queues empty. Ignores dependency maps to maximize throughput. Post-run integration fixes merge parallel outputs, adds or adjusts tests, runs tests, and fixes failures. Every launch must use agent_type sparky; other roles are invalid. Emphasizes continuous project movement with maximum path context per subagent.1.1kinstalls17Parallelparallel is a skill that acts as an orchestrator for subagents, triggered by explicit /parallel-task commands. It parses a markdown plan file, infers a task dependency graph, and delegates tasks to parallel subagents in waves until all are complete, updating the plan with logs after each task. It enforces TDD RED-to-GREEN test evidence or a documented non-testable alternative, and requires each task to be committed locally before the next wave. Developers use it to execute multi-task plans concurrently.1installs18Create Hookcreate-hook is a skill that creates or updates Codex hooks at global or project scope. It asks for scope and hook type when ambiguous, then writes the hook config, scripts, and any required feature-flag config across five event types (SessionStart, UserPromptSubmit, PreToolUse, PostToolUse, Stop). A developer uses it to automate deterministic behavior around Codex tool use, such as prevention, logging, or prompt shaping.0installs19Img To Frontendimg-to-frontend is a skill that runs an image-first design-to-code workflow for websites. It first generates four distinct concept images, pauses for the user to select one, converts the selected image into a granular implementation prompt, then builds the real page and iterates against screenshots until structure, typography, spacing, color, and responsive behavior match the reference. A developer uses it to turn a visual design into a near-exact frontend implementation.0installs20Pluginstallerpluginstaller is a skill that adds a Codex plugin from a GitHub repo to a repo or personal marketplace using the current Codex plugin layout. It locates the plugin root via .codex-plugin/plugin.json, validates the manifest, copies the full plugin bundle to a marketplace-visible location, and creates or updates the marketplace file with the documented schema. It then restarts Codex and verifies the plugin appears in /plugins. Developers use it to install Codex plugins correctly.0installs