What problem it solves

Developers need repeatable workflows for researching changes, making design decisions, implementing work, explaining topics, and preparing pull requests for review. The project packages these workflows as installable Claude Code skills that can also be invoked with specific tasks or phases.

Key features

• Structured research and implementation — The rpi skill supports research, PRD followed by TDD, technical design, or research followed by implementation. Each phase works in a fresh session by default.

  Source: README

• Concise technical explanations — The show-me skill explains the current topic with diagrams, code-shape sketches, and focused HTML artifacts.

  Source: README

• Visual pull request outlines — The visual-pr skill creates or updates a pull request with a concise visual outline that helps reviewers understand the change.

  Source: README

• Instruction-focused CLAUDE.md rewriting — The improve-claude-md skill rewrites CLAUDE.md using `<important if>` blocks to improve instruction adherence.

  Source: README

• Repo-local agentic workflows — The build-iterated-agentic-loop skill creates a repo-local skill, an iterated coding-agent GitHub Actions workflow, and supporting prompt, memory, and reference files. The design-control-loop skill interviews developers to design an agentic control loop and builds it as locally runnable components plus a scheduled workflow.

  Source: README

Quick start

Copied verbatim from the project's README.

npx skills add humanlayer/skills --skill rpi

/rpi research how the collections connect to the database
/rpi make a PRD
/rpi what's next .agents/artifacts/my-change
/rpi implement .agents/artifacts/my-change

/show-me

Who it may suit

This project may suit developers using Claude Code or Codex who want task-oriented workflows for research, design, implementation, explanation, and pull request preparation.

Topics to discuss

• Which additional Claude Code skills would be most useful for repository-based development workflows?

• How should skills handle local documentation, writable task directories, and session boundaries?

• What principles should guide the design of repo-local agentic loops and scheduled coding-agent workflows?

Written by ReadmeX AI from the project's README, source and public GitHub data; it may contain mistakes.

Read the original README https://github.com/humanlayer/skills#readme

ReadmeX opened this community for a trending project; it is not run by the project. Maintainers can claim it from the community page.