Turns a codebase, its docs, schemas and PDFs into a queryable knowledge graph for AI coding assistants, built locally with tree-sitter AST parsing.
DeusData/codebase-memory-mcp
An MCP server that indexes code, docs and infrastructure into a local knowledge graph so AI coding agents can query structure instead of re-reading files.
- Stars
- 47,045
- Forks
- 3,863
- Open issues
- 622
- Last commit
- 2 hours ago
What is it?
codebase-memory-mcp is a native C MCP server that builds a persistent local knowledge graph from a repository. It parses source code with vendored tree-sitter grammars, links symbols through calls, imports and inheritance, and connects them to Markdown and PDF documentation as well as Docker, Kubernetes, Helm, Terraform, Ansible and CI configuration. Coding agents can then ask for callers, dependencies, change impact or deployment paths through its 17 MCP tools, or query the graph directly from the CLI.
Indexing and queries run entirely on the local machine, with no hosted service or API key required. The included semantic search uses an embedding model compiled into the binary. The installer configures supported agent clients such as Claude Code, Codex, Cursor and Windsurf, adding instructions, skills and hooks where each client supports them.
CBM Lens is an optional browser interface in open beta that visualizes the graph as architecture maps, a 3D call-graph view, and source readers with coverage diagnostics. It also includes a local chat that can answer questions about selected code using an on-device model.
Key features
- 01Exposes 17 MCP tools for structural search, call tracing, architecture summaries and change impact.
- 02Parses many languages with vendored tree-sitter grammars and resolves calls across files.
- 03Links Markdown, PDF and ADR documentation to the code symbols they reference.
- 04Maps Docker, Kubernetes, Helm, Terraform and Ansible relationships to source.
- 05Runs semantic, BM25 and structural search without an external API or service.
- 06Installs MCP configuration and agent skills for 45 supported client integrations.
Use cases
- Helping an AI coding agent navigate an unfamiliar large codebase without re-reading many files.
- Tracing which services and images a Kubernetes or Terraform change reaches.
- Estimating which symbols and tests a set of uncommitted changes could affect.
- Exploring a repository's architecture visually with the CBM Lens interface.
Topics
FAQ
- What is codebase-memory-mcp?
- An MCP server that indexes code, docs and infrastructure into a local knowledge graph so AI coding agents can query structure instead of re-reading files. codebase-memory-mcp is a native C MCP server that builds a persistent local knowledge graph from a repository. It parses source code with vendored tree-sitter grammars, links symbols through calls, imports and inheritance, and connects them to Markdown and PDF documentation as well as Docker, Kubernetes, Helm, Terraform, Ansible and CI configuration. Coding agents can then ask for callers, dependencies, change impact or deployment paths through its 17 MCP tools, or query the graph directly from the CLI.
- Is codebase-memory-mcp free to use? What license does it have?
- codebase-memory-mcp is open source under the MIT license. Check the license terms before using it in commercial products.
- Is codebase-memory-mcp still maintained?
- Actively maintained. The last commit to codebase-memory-mcp was on Oct 21, 2026.
- What can I use codebase-memory-mcp for?
- Helping an AI coding agent navigate an unfamiliar large codebase without re-reading many files. Tracing which services and images a Kubernetes or Terraform change reaches. Estimating which symbols and tests a set of uncommitted changes could affect. Exploring a repository's architecture visually with the CBM Lens interface.
- Where can I get codebase-memory-mcp?
- The source code and installation instructions are on GitHub: https://github.com/DeusData/codebase-memory-mcp
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
Indexes a codebase into a knowledge graph of dependencies, call chains and execution flows, and serves it to AI coding agents via MCP tools.
Indexes multi-language codebases into a knowledge graph so you can query, search, and edit code in plain English via an AI agent or MCP server.
A local code knowledge graph that gives AI coding tools precise review context over MCP and a CLI.
A local catalog and control plane that lets Codex or Claude search 2,700+ SKILL.md playbooks, pick exact skills, validate a stack and preview an install plan.
Open-source AI memory platform that turns documents, code and conversations into a self-hosted knowledge graph for AI agents.