
Repomix Codebase Packer
Packs a code repository into one AI-readable file so the agent can explore and search it.
- Repository
- yamadashy/repomix
- Stars
- 29,118
- Licence
- MIT
- Files
- 1 file
What it does
This skill teaches the agent to use the Repomix command-line tool, which bundles a whole code project (local folder or a GitHub repository) into a single structured file. The agent then searches that file for structure, functions, routes, or models and reports what it finds, along with size and token-count metrics. It is useful when you want an AI to understand a codebase without opening every file yourself.
Great for
- Developers who want a quick overview of an unfamiliar codebase
- Anyone preparing a project's full source as context for an AI analysis
- Engineers estimating how many tokens a repository would use
Skip it if
- Very sensitive or private code you do not want sent over the network: the remote mode fetches from GitHub, and packing creates a copy of your code that the agent then reads, so use local-only workflows instead.
- Tasks needing guaranteed secret removal: the skill relies on Repomix's built-in checks and tells the agent to trust them, which may miss some credentials.
Try asking
- “Analyze this repo and tell me its main components: yamadashy/repomix”
- “How many tokens is this project?”
- “Find all authentication-related code in ./src”
- “Pack facebook/react with compression and show me the API routes”
Install in 3 steps
- 1
Download the skill
The ZIP contains the 1 original files, the MIT licence and an attribution note.
clawhub.zip - 2
Add it to your agent
Claude Code (all projects)
mkdir -p ~/.claude/skills && unzip clawhub.zip -d ~/.claude/skills/Claude Code / Agent SDK (this project only)
mkdir -p .claude/skills && unzip clawhub.zip -d .claude/skills/Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.
- 3
Ask for the task
No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.
Get better results
- Name a specific repository or folder so the agent packs only what you need.
- Use the include option to limit the scope to certain file types, which keeps output small.
- For large repositories, ask for compression to reduce token usage.
What's inside
- SKILL.md
Redistributed unmodified from yamadashy/repomix under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.