RepoBrain
Gives the agent an understanding of your repository: it indexes the code and answers questions about it in Claude Code, Cursor, Codex, Windsurf and more
Medium risk
We rate an entry medium when the tool runs code, makes network calls or reads project files. Check what exactly it does before installing.
Why this level
- Reads repository files during indexing
- Sends code to an external model via a key
Install
In your terminal, with SkillFoxx CLI
npx skillfoxx add plugins/repobrainDetects the agents on your machine, checks the risk and pins the version.
Other ways to install
Run one by one in the Claude Code chat
/plugin marketplace add study8677/repobrain
/plugin install repobrain@repobrainInstall the plugin: /plugin marketplace add study8677/repobrain, then /plugin install repobrain@repobrain. The rb CLI and engine auto-install on first run.
Other ways from the author
/plugin marketplace add study8677/repobrain
/plugin install repobrain@repobrainThe rb CLI and engine auto-install on first run.
This is third-party code. Review the repository files before installing.
What it does
RepoBrain builds an index of the codebase and lets you ask questions about it right from the agent. The plugin installs an engine and an rb CLI alongside it, picked up on first run. It can connect as an MCP server through rb-mcp and works across several agents and IDEs. Analysis needs model access via a key, and different providers are supported.
Who it is for. For developers who need to quickly understand an unfamiliar or large repository.
Good fit when
- You need to understand a large or unfamiliar codebase
- You need to find where a feature lives and how parts connect
- You want to ask questions about the repo in natural language
Not a fit when
- The repository is small and clear without an index
- You cannot send code through an external model key
Example request
Index this repository with RepoBrain and explain how the auth module is builtLimitations
Analysis needs a model key: with OpenAI, access from Russia without VPN does not work, though other providers are supported. Indexing reads the repository files.
How to disable. Remove the repobrain plugin via /plugin and uninstall the rb packages via your Python package manager.
Security check
- Reads repository files during indexing
- Sends code to an external model via a key
README in short
The README describes RepoBrain as a way to ask questions about your codebase across agents and IDEs. The recommended install is a plugin in Claude Code and Codex, with the rb CLI and engine installed automatically. It also shows pipx installation and connecting as an MCP server via rb-mcp. A model key set in an environment variable is required.
FAQ
How do I connect it as MCP?
With claude mcp add repobrain rb-mcp -- --workspace $(pwd) after the engine is installed.
Which key is needed?
A model key such as OPENAI_API_KEY; other compatible providers are also supported.
Related
A skills library that gives coding agents a development process: brainstorming, planning, TDD, subagents and code review
Skills for real engineers by Matt Pocock
Skills For Real Engineers
Small composable skills for engineering with agents: plan grilling, TDD, bug diagnosis, code review and architecture
GitHub toolkit for spec-driven development: the specify CLI adds agent commands and skills to a project, from principles to implementation
Reference MCP servers
Model Context Protocol servers
Official reference MCP servers: Filesystem, Fetch, Git, Memory, Sequential Thinking, Time and Everything