</>MCP Agents Market
Agent

codebase-memory-mcp

by DeusData39.2kCUpdated 2026-08-17

High-performance code intelligence MCP server. Indexes codebases into a persistent knowledge graph — average repo in milliseconds. 158 languages, sub-ms queries, 99% fewer tokens. Single static binary, zero dependencies.

Claude CodeCodex

codebase-memory-mcp is a high-performance code intelligence AI sub-agent that indexes codebases into persistent knowledge graphs. Built in pure C as a single static binary with zero dependencies, it supports 158 programming languages and delivers sub-millisecond query responses. The tool parses entire repositories in milliseconds using tree-sitter AST analysis, reducing token consumption by 99% compared to file-by-file exploration while providing structural queries, call graphs, and cross-service linking for AI coding agents.

Key Features

Extreme indexing speed: processes the Linux kernel (28M lines, 75K files) in 3 minutes with RAM-first pipeline and in-memory SQLite
158 vendored tree-sitter grammars compiled into the binary with Hybrid LSP semantic type resolution for 10 major languages
15 MCP tools including architecture overview, semantic search with bundled embeddings, BM25 full-text search, call graph tracing, and dead code detection
Cross-service linking for HTTP, gRPC, GraphQL, and tRPC with route-to-call-site matching and pub-sub channel detection across 8 languages
Built-in 3D graph visualization UI served from the binary at localhost with multi-galaxy layout for cross-repo intelligence
Team-shared graph artifacts: commit a single zstd-compressed snapshot to skip reindexing for teammates
Automatic configuration for 43 supported client surfaces with background file watching and git diff impact mapping
100% local processing with no API keys, language runtimes, or hosted services required

Use Cases

  • 01Enable AI coding agents to navigate large codebases efficiently by querying structural relationships instead of reading entire files
  • 02Trace function call chains and identify dead code across multi-file projects with import-aware analysis
  • 03Map uncommitted git changes to affected symbols with automated risk classification for impact analysis
  • 04Discover architectural boundaries and functional modules through Louvain community detection on call graphs
  • 05Link HTTP routes to their implementations and track cross-service dependencies in microservice architectures
  • 06Share compressed knowledge graph snapshots with team members to avoid redundant indexing on cloned repositories

Related Agents

View more

unsloth

Local UI to run and train LLMs and diffusion models, including Qwen3.8, Kimi K3, MiniMax-H3, Gemma 4, DeepSeek-V4, FLUX and more.

Agent72.6k

codegraph

Pre-indexed code knowledge graph, auto syncs on code changes, for Claude Code, Codex, Gemini, Cursor, OpenCode, AntiGravity, Kiro, and Hermes Agent — fewer tokens, fewer tool calls, 100% local

Agent66.6k

claude-code-best-practice

from vibe coding to agentic engineering - practice makes claude perfect

Agent64.6k

system_prompts_leaks

Extracted system prompts from Anthropic - Claude Fable 5, Opus 5, Claude Design, Claude Code. OpenAI - ChatGPT GPT-5.6-Sol, Codex. Google - Gemini 3.5 Flash, 3.1 Pro, Antigravity. xAI - Grok, Cursor, Copilot, VS Code, Perplexity, and more. Updated regularly.

Agent63k

codebase-memory-mcp — FAQ

What is codebase-memory-mcp?+

codebase-memory-mcp is an AI sub-agent that builds persistent knowledge graphs from source code, enabling AI coding assistants to answer structural queries about functions, classes, call chains, and cross-service dependencies. It runs entirely locally with no external services or API keys required.

How do I install codebase-memory-mcp?+

Run the one-line installer for macOS/Linux: curl -fsSL https://raw.githubusercontent.com/DeusData/codebase-memory-mcp/main/install.sh | bash. On Windows, download install.ps1, unblock it with Unblock-File, and execute it in PowerShell. The installer auto-detects and configures supported AI clients.

Which AI clients work with codebase-memory-mcp?+

The tool automatically configures 43 supported client surfaces including Claude Code, Codex, and other MCP-compatible agents. The installer detects installed clients and sets up their MCP configuration entries, skills, and lifecycle hooks where supported.

Does codebase-memory-mcp require API keys or dependencies?+

No. codebase-memory-mcp ships as a single native executable with all tree-sitter grammars and embedding models compiled in. It requires no language runtimes, Docker, hosted services, or API keys—processing happens 100% locally.

Is codebase-memory-mcp free to use?+

Yes, codebase-memory-mcp is open source and released under the MIT license. All code and release binaries are available on GitHub with no usage fees or subscription costs.

How does codebase-memory-mcp reduce token usage?+

By building a knowledge graph, the tool answers structural queries (like finding all callers of a function) in one operation consuming ~3,400 tokens, versus ~412,000 tokens required for file-by-file grep exploration—a 99% reduction.

How do I install codebase-memory-mcp?+

Open the source repository on GitHub and follow its README. codebase-memory-mcp is a agent — MCP Agents Market links you directly to the official repo.

Is codebase-memory-mcp free?+

codebase-memory-mcp is an open-source project hosted on GitHub. Check the repository for its license and any usage requirements.

Related searches

Developer Tools AgentsCode Review AgentsKnowledge & Memory AgentsBest MCP ServersBest SkillsBest PluginsTop rated listingsClaude Code toolsBrowse all categories