HomeProjectsArticlesPK BoardAbout
Explore Projects

Codebase Memory MCP

Active
GitHub C MIT

Description

High-performance code intelligence MCP server that indexes codebases into a persistent knowledge graph, supporting 66 languages with sub-millisecond queries and 99% fewer tokens.

Key Features

  • Extreme indexing speed: Linux kernel (28M LOC) indexed in 3 minutes via RAM-first pipeline
  • 159 languages supported through vendored tree-sitter grammars in a single binary
  • Hybrid LSP semantic type resolution for Python, TypeScript, PHP, C#, Go, C, and C++
  • 120x fewer tokens compared to file-by-file exploration (~3,400 vs ~412,000 tokens)
  • 14 MCP tools: search, trace, architecture analysis, impact analysis, dead code detection
  • Plug and play: single static binary for macOS/Linux/Windows, no Docker or runtime dependencies

Use Cases

💡 Accelerating AI coding agents with deep codebase understanding
💡 Performing impact analysis before making code changes
💡 Detecting dead code and unused dependencies across large repositories
💡 Tracing function call chains and cross-service HTTP routes
💡 Building persistent knowledge graphs for ongoing project documentation

Strengths & Limitations

✅ Strengths

  • • Actively maintained, recent updates
  • • High community interest (41.9k stars)
  • • Permissive open-source license (MIT)

Tags

mcp knowledge-graph code-analysis tree-sitter code-intelligence

Categories

🧩 Agent Memory 🔌 MCP Tools

Quick Start

1. Run the one-line installer: curl -fsSL https://raw.githubusercontent.com/DeusData/codebase-memory-mcp/main/install.sh | bash. 2. Restart your coding agent (Claude Code, Codex CLI, Gemini CLI, etc.). 3. Say 'Index this project' — the server indexes automatically. 4. For graph visualization, add --ui flag and open localhost:9749. 5. Use 'codebase-memory-mcp update' to keep the binary up to date.
Visit GitHub Visit Website View Docs

Related Projects

Headroom

68.6k · Python
Active A+

Context compression layer that compresses tool outputs, logs, files, and RAG chunks before they reach the LLM, saving 60-95% of tokens while preserving answer quality. Available as library, proxy, or MCP server.

token-compressionoptimizationmcp +2
  • · Compress everything before LLM receives it, reducing 60-95% of tokens
  • · Three usage modes: library, proxy, and one-command agent wrapping
  • · Built-in MCP server with cross-agent shared memory
View full details → View on GitHub

MCP Memory Service

1.9k · Python
Active A+

Open-source persistent memory service for AI agents, supporting LangGraph, CrewAI, and AutoGen with REST API, knowledge graph, and autonomous memory consolidation.

agent-memorymcpknowledge-graph +2
  • · Open-source persistent memory service for AI agents — provides long-term memory storage and retrieval via REST API for any MCP-compatible framework
  • · Knowledge graph-backed memory with autonomous consolidation — agents organize and merge memories into structured knowledge representations over time
  • · Multi-framework compatibility — works with LangGraph, CrewAI, AutoGen, and other popular agent frameworks out of the box
View on GitHub

code-review-graph

31.1k · Python
Active A+

Local-first code intelligence graph for MCP and CLI. Builds a persistent map of the codebase so AI coding tools read only what matters, with benchmarked context reductions on code review and large-repo workflows.

code-reviewknowledge-graphmcp +3
  • · Local-first — all indexing happens on the developer's machine, no data leaves the host
  • · Code intelligence graph — parses code with tree-sitter and builds an entity-relation graph
  • · MCP integration — exposes Model Context Protocol endpoints for Claude Code, Cursor, and other agents
View full details → View on GitHub

Related Articles

MCPModel Context ProtocolClaude Code

Best MCP Servers in 2026: 25 Servers Worth Integrating, From Official to Community

A 2026 review of the 25 Model Context Protocol (MCP) servers most worth integrating, covering GitHub, databases, browsers, filesystem, code search, Google Workspace, Slack, Notion, and more. Includes official SDK vs community implementations and configuration in Claude Code, Cursor, and Cline.

Read more →