git-mcp-server and GitIntel-MCP-Server

git-mcp-server
64
Established
GitIntel-MCP-Server
42
Emerging
Maintenance 13/25
Adoption 11/25
Maturity 18/25
Community 22/25
Maintenance 13/25
Adoption 4/25
Maturity 9/25
Community 16/25
Stars: 192
Forks: 47
Downloads:
Commits (30d): 0
Language: TypeScript
License: Apache-2.0
Stars: 8
Forks: 6
Downloads:
Commits (30d): 0
Language: TypeScript
License: MIT
No risk flags
No Package No Dependents

About git-mcp-server

cyanheads/git-mcp-server

An MCP (Model Context Protocol) server enabling LLMs and AI agents to interact with Git repositories. Provides tools for comprehensive Git operations including clone, commit, branch, diff, log, status, push, pull, merge, rebase, worktree, tag management, and more, via the MCP standard. STDIO & HTTP.

Provides 28 tools organized by workflow (staging, branching, remote ops, advanced), plus session-based resource management and a structured wrap-up prompt for guided git workflows. Built on pluggable architecture with declarative tool definitions, dependency injection via tsyringe, and support for commit signing—with cross-runtime detection for Bun and Node.js via auto-spawned git processes. Integrates with MCP clients (Cline, Claude Desktop) over stdio or HTTP transports, with configurable git identity, path sanitization for security, and optional OpenTelemetry observability.

About GitIntel-MCP-Server

hoangsonww/GitIntel-MCP-Server

A Git intelligence MCP server built with Node.js and TypeScript that analyzes local Git repositories to surface insights like hotspots, churn, temporal coupling, knowledge maps, and risk scoring. Designed for AI agents, it exposes repository analytics tools through the Model Context Protocol (MCP) while keeping all data local and read-only.

Exposes 12 specialized analysis tools (hotspots, temporal coupling, knowledge maps, churn, complexity trends, risk scoring) that transform raw git output through multi-stage pipelines into scored, formatted tables and actionable insights. Built on stdio JSON-RPC transport with safe subprocess execution via `execFile`, it auto-detects repositories from working directory or accepts explicit paths, and ships with pre-computed resources (repository summary, activity feeds) for quick snapshot access. Designed for local-first deployment across Claude Code, Codex, and other MCP clients—includes Docker multi-stage builds, Kubernetes Kustomize manifests, and Terraform/CloudFormation templates for containerized or cloud-native environments.

Scores updated daily from GitHub, PyPI, and npm data. How scores work