dgalarza/claude-code-workflows
9 skills329 installs531 starsGitHub
Install
npx skills add https://github.com/dgalarza/claude-code-workflowsSkills in this repo
1Process Meeting TranscriptProcess Meeting Transcript is a Claude skill that converts raw meeting transcripts from Granola or similar sources into well-structured Obsidian notes - complete with YAML frontmatter, extracted action items, a summary and a cleanly formatted transcript. A solo builder reaches for it after a call to capture decisions and follow-ups without re-reading the whole recording.75installs2Conventional CommitsConventional Commits is a Claude skill that guides you to write Git commit messages following the Conventional Commits v1.0.0 specification - the right type, optional scope, description, body and footer. A solo builder reaches for it when committing code or reviewing PR history to keep an explicit, tool-parseable commit log that powers automated changelogs and semantic versioning.56installs3Tdd WorkflowTDD Workflow is a Claude skill that enforces a disciplined test-driven development loop: write the simplest failing test for one piece of behavior, confirm it fails, write just enough code to pass, confirm it passes, then refactor. It deliberately avoids writing whole test files up front or building multiple features at once. A solo builder reaches for it to keep implementation honest and incremental.43installs4Linear ImplementLinear Implement is a Claude skill that takes a Linear issue ID and runs an end-to-end implementation workflow: it fetches the issue, plans the work, builds with test-driven development, launches parallel security and Rails best-practice reviews, applies the feedback, and opens a PR wired back to Linear. A solo Rails builder reaches for it to go from ticket to reviewed pull request without manually stitching the steps together.35installs5Codebase Readinesscodebase-readiness runs an Agent-Ready Codebase Assessment, scoring a repo across 8 dimensions with parallel agents to produce a weighted 0-100 score, a band rating, and an improvement roadmap, framed against the benchmark of high-volume AI-generated pull requests. It supports Ruby, Python, PHP, TypeScript, JavaScript, Go, Java, Scala, and Rust. A solo builder reaches for it to find out how well their codebase supports autonomous AI agent work and what to fix first.32installs6Parallel Code ReviewParallel Code Review is a Claude skill that fans out several specialized review agents - security, architecture, performance - across a changeset at the same time, then consolidates their findings into one report while tracking decisions to prevent overlap. A solo builder reaches for it before merging a large PR to get thorough, multi-perspective review in the time of a single pass.28installs7Rspec TestingRSpec Testing is a Claude skill for writing, reviewing and improving RSpec test suites in Ruby on Rails applications. It applies Better Specs and thoughtbot guidance across model, controller, system, component, service and integration specs, emphasizing clear Arrange-Act-Assert structure and TDD. A Rails builder reaches for it to raise test coverage and quality without memorizing every RSpec convention.25installs8Gridfinity Baseplate PlannerGridfinity Baseplate Planner is a Claude skill for planning gridfinity baseplates optimized for 3D printing. It converts drawer or shelf measurements to millimeters, calculates the optimal grid, slices large layouts into printer-bed-sized chunks, and works out padding for imperfect fits before you generate STLs at gridfinity.perplexinglabs.com. A maker reaches for it to organize drawers without hand-calculating grid math.20installs9Agent Readyagent-ready scaffolds the documentation and structural artifacts that make a codebase legible to AI agents - AGENTS.md, ARCHITECTURE.md, and a docs/ tree - following progressive-disclosure patterns. It analyzes the codebase, generates the artifacts, and audits existing ones for staleness and coherence. As the remediation companion to codebase-readiness, a solo builder reaches for it to actually build the docs an agent needs rather than just score the gaps.15installs