octocode
Code research platform for AI agents; find, understand, and prove context across your code and all of GitHub, in a fraction of the tokens. One toolset, MCP or C
What is octocode?
Code research platform for AI agents; find, understand, and prove context across your code and all of GitHub, in a fraction of the tokens. One toolset, MCP or CLI
Overview of octocode
What it is
Octocode is an evidence-first code research platform for AI agents and developers. It researches local code and external code (GitHub repositories, PRs, npm) with one toolset: ripgrep + AST search, trees, precise reads, and LSP. It can be used as a CLI or MCP server, backed by a Rust engine for fast, token-efficient results across single files or mega-repos.
Key features
- Agent-driven, efficient flows: chains cheap steps into an optimized research flow with smart pagination and minification, so the model never over-fetches.
- Scales to monorepos: spot a pattern in one repository, follow the PR that introduced it, then trace it across other repositories and local files.
- Smart GitHub flow: parallel bulk queries across code, PRs, commits, issues, and repositories with the same search-broad, read-narrow discipline.
- Works without GitHub: clone any repository and point local tools (search, AST, LSP, content) at it.
- Reads shape, not noise: on-the-fly minify/skeletonize across 70+ languages.
- Fast, self-contained: search, parsing, navigation, and redaction run in one prebuilt Rust engine.
- Safe by default: every byte to the model is scanned and secrets redacted first.
Use cases
- Codebase questions: search local or GitHub code, read exact regions, browse trees, and carry file/line anchors into the answer.
- Implementation research: compare patterns across repositories, npm packages, pull requests, commits, and local files before changing code.
- Semantic navigation: resolve definitions, references, callers/callees, call hierarchy, hovers, symbols, diagnostics, and type relationships through LSP.
- Structural matching: run AST-shaped searches with patterns or YAML rules so comments and strings do not become false positives.
- Large-file context: minify, skeletonize, or paginate code so agents spend tokens on relevant structure instead of boilerplate.
- Agent workflows: same engine through MCP, CLI, and Agent Skills.
Getting started
Prerequisites: Node.js 20.12+ Run the Octocode CLI with npx: npx octocode --help Authenticate with GitHub (optional, but unlocks private repositories and higher API rate limits): npx octocode auth login npx octocode status Choose your interface: CLI: npx octocode MCP: add to your MCP client config (or use one-click install options in the repository README).
License and pricing
The license is MIT. The project is open source.
Based on the project's public README and repository metadata.