codebase-memory-mcp

Local code intelligence for developers and AI agents. Connect code, docs and infrastructure in a persistent graph, with CBM Lens, MCP tools, agent integrations

What is codebase-memory-mcp?

Local code intelligence for developers and AI agents. Connect code, docs and infrastructure in a persistent graph, with CBM Lens, MCP tools, agent integrations and change-aware test selection.

Overview of codebase-memory-mcp

What it is

codebase-memory-mcp is a high-performance code intelligence MCP server. It indexes codebases into a persistent knowledge graph using tree-sitter AST analysis across 162 languages, enhanced with Hybrid LSP semantic type resolution for Python, TypeScript, JavaScript, JSX, TSX, PHP, C#, Go, C, C++, Java, Kotlin, Rust, and Perl. The tool ships as a single static binary with zero dependencies for macOS, Linux, and Windows. All processing happens locally; no API key or hosted service is required. It provides 17 MCP tools and a built-in 3D graph visualization at localhost:9749. The design and benchmarks are described in the preprint Codebase-Memory: Tree-Sitter-Based Knowledge Graphs for LLM Code Exploration via MCP (arXiv:2603.27277).

Key features

  • Indexes an average repository in milliseconds; the Linux kernel (28M LOC, 75K files) is indexed in 3 minutes.
  • Answers structural queries in under 1ms.
  • Uses 120x fewer tokens than file-by-file exploration: 5 structural queries use ~3,400 tokens vs ~412,000.
  • Supports 162 languages via vendored tree-sitter grammars compiled into the binary.
  • Includes 17 MCP tools for search, trace, architecture, impact analysis, dead code detection, cross-service HTTP linking, ADR management, and more.
  • Provides CBM Atlas, a built-in human interface at localhost:9749 with a reading IDE, 3D galaxy visualization, and source-cited answers.
  • Indexes infrastructure-as-code: Dockerfiles, Kubernetes manifests, and Kustomize overlays as graph nodes with cross-references.
  • Features a Session Coordination Daemon that shares one per-account daemon across multiple clients (Claude Code, Codex, OpenCode, etc.) and manages long-lived background services.

Use cases

  • AI coding agents exploring codebases with structural queries.
  • Impact analysis and dead code detection.
  • Cross-service HTTP linking and ADR management.
  • Infrastructure-as-code indexing for Dockerfiles, Kubernetes manifests, and Kustomize overlays.
  • Code search, trace, and architecture analysis via MCP tools.
  • Human-readable code exploration through CBM Atlas.

Getting started

One-line install (macOS / Linux): bash curl -fsSL https://raw.githubusercontent.com/DeusData/codebase-memory-mcp/main/install.sh | bash Windows (PowerShell): powershell Invoke-WebRequest -Uri https://raw.githubusercontent.com/DeusData/codebase-memory-mcp/main/install.ps1 -OutFile install.ps1 Unblock-File .install.ps1 .install.ps1 After installation, restart your coding agent and say “Index this project”. Manual install instructions are available in the repository README.

License and pricing

MIT license. Open source.

Based on the project's public README and repository metadata.

Add this badge to your README

Featured on YunoDB Badge styles & help

Maintain this project? Show that it is featured on YunoDB.

Markdown[![Featured on YunoDB](https://yunodb.com/badge/deusdatacodebase-memory-mcp.svg)](https://yunodb.com/tools/deusdatacodebase-memory-mcp?utm_source=github&utm_medium=badge)
HTML<a href="https://yunodb.com/tools/deusdatacodebase-memory-mcp?utm_source=github&utm_medium=badge"><img src="https://yunodb.com/badge/deusdatacodebase-memory-mcp.svg" alt="Featured on YunoDB" height="20"></a>