bgauryy/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

What it solves

Octocode is an evidence-first code research platform designed to provide AI agents and developers with precise, token-efficient context from both local filesystems and external sources like GitHub and npm. It prevents agents from "guessing" by allowing them to chain search, structural analysis, and semantic navigation to find the exact code regions needed for a task.

How it works

Octocode uses a high-performance Rust engine to power a suite of 17 tools. It can be deployed as a Model Context Protocol (MCP) server for AI assistants or as a standalone CLI. The system integrates ripgrep, AST (Abstract Syntax Tree) search, and Language Server Protocol (LSP) to perform deep code analysis. It supports a "search-broad, read-narrow" workflow where agents can find a pattern, fetch only the specific matched lines, and use minification/skeletonization to reduce token usage while preserving structural meaning across 70+ languages.

Who it’s for

  • AI Agent Developers: Those building agents that need to research and modify codebases without over-fetching tokens.
  • Software Engineers: Developers who want a powerful, unified CLI for researching local and remote repositories. -- AI Assistant Users: Users of MCP-compatible clients (like Cursor, VS Code, Windsurf) who want their AI to have deep, evidence-based access to their code.

Highlights

  • Unified Research Flow: Single toolset for local code, GitHub repositories, PRs, issues, and npm packages.
  • Token Efficiency: Built-in minification and skeletonization to strip boilerplate and reduce model context window pressure.
  • Semantic Navigation: Full LSP support for resolving definitions, references, callers, and type relationships.
  • Structural Search: AST-based searching to avoid false positives from comments or strings.
  • High Performance: Powered by a Rust engine for fast execution across single files or massive monorepos.
  • MCP Integration: Native support for the Model Context Protocol for one-click installation in modern AI IDEs.

Related

  • Project
    Muvon/octocode

    Octocode is an open‑source Rust tool that builds a live, tree‑sitter‑derived knowledge graph of a codebase and exposes it via an MCP server. AI assistants can perform semantic search, LSP‑level navigation, and graph‑based queries (imports, calls, etc.) across 16 languages. It supports optional vector embeddings for hybrid retrieval, offers a local‑first privacy model, and includes a reproducible benchmark showing strong code‑search performance.

  • Project
    yc-duan/fastctx

    A high-performance Rust-based MCP server that provides structured repository tools for AI agents to efficiently read, search, and edit codebases.

  • Project
    cocoindex-io/cocoindex-code

    An AST-based semantic code search tool that enables AI coding agents and developers to find code by meaning and structure, reducing token usage by up to 70%.

  • Project
    Context-Engine-AI/Context-Engine

    A semantic code search and symbol intelligence system that provides AI coding assistants with deep codebase context, memory, and navigation tools via MCP.

  • Project
    giancarloerra/SocratiCode

    An open-source codebase context engine that gives AI assistants deep semantic understanding of large repositories through hybrid search and dependency graphs.