MCP Git Ingest
@adhikasp
About MCP Git Ingest
A Model Context Protocol (MCP) server that helps read GitHub repository structure and important files.
Config
Add this server to your MCP-compatible client using the configuration below.
{
"mcpServers": {
"mcp-git-ingest": {
"command": "uvx",
"args": [
"--from",
"git+https://github.com/adhikasp/mcp-git-ingest",
"mcp-git-ingest"
]
}
}
}Tools
No tools detected
We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.
Overview
What is MCP Git Ingest?
MCP Git Ingest is a Model Context Protocol (MCP) server that reads GitHub repository structure and important files. It provides two tools: github_directory_structure for a tree-like directory view and github_read_important_files for reading specified file contents. It is inspired by gitingest.com and is intended for developers who need to programmatically explore repositories via an MCP client.
How to use MCP Git Ingest?
Configure the server in your MCP client’s mcpServers block using uvx with the GitHub URL. Then invoke the tools via an MCP client (e.g., mcp-client-cli). The server exposes github_directory_structure(repo_url) and github_read_important_files(repo_url, file_paths).
Key features of MCP Git Ingest
- Provides
github_directory_structureandgithub_read_important_filestools. - Uses
fastmcpfor MCP server creation. - Uses
gitpythonfor Git repository operations. - Creates deterministic temporary directories based on repo URL hash.
- Cleans up temporary files after each operation.
- Generates Unicode tree representation of directory structure.
Use cases of MCP Git Ingest
- Quickly visualize the directory layout of any public GitHub repository.
- Read key files (README, config, source code) from a repository for analysis.
- Integrate repository exploration into AI‑assisted coding workflows.
- Automate code review preparation by fetching project structure and important files.
FAQ from MCP Git Ingest
What tools does MCP Git Ingest provide?
It provides two tools: github_directory_structure to return a tree‑like directory layout, and github_read_important_files to return the contents of specified files in a repository.
What are the dependencies and runtime requirements?
It requires Python 3.8+, fastmcp, and gitpython. The server is run via uvx from the git repository.
How does MCP Git Ingest handle temporary repository data?
It clones each repository into a deterministic temporary directory named by hashing the repository URL. After processing, it removes the temporary directory using a finally block to avoid file accumulation.
Where does my data live?
All repository data is cloned into temporary directories on the local machine and is deleted after each tool call. No data is sent to external services beyond the initial GitHub clone.
What transport does MCP Git Ingest use?
The README does not specify a transport; it is assumed to use the standard MCP transport (stdio) as is typical for fastmcp‑based servers.
Frequently asked questions
What tools does MCP Git Ingest provide?
It provides two tools: `github_directory_structure` to return a tree‑like directory layout, and `github_read_important_files` to return the contents of specified files in a repository.
What are the dependencies and runtime requirements?
It requires Python 3.8+, `fastmcp`, and `gitpython`. The server is run via `uvx` from the git repository.
How does MCP Git Ingest handle temporary repository data?
It clones each repository into a deterministic temporary directory named by hashing the repository URL. After processing, it removes the temporary directory using a `finally` block to avoid file accumulation.
Where does my data live?
All repository data is cloned into temporary directories on the local machine and is deleted after each tool call. No data is sent to external services beyond the initial GitHub clone.
What transport does MCP Git Ingest use?
The README does not specify a transport; it is assumed to use the standard MCP transport (stdio) as is typical for `fastmcp`‑based servers.
Basic information
More Version Control MCP servers
Integrate MCP with GitHub Copilot
skillsLearn how to use MCP Servers with GitHub Copilot
MCP Atlassian Server (by phuc-nt)
phuc-ntMCP server connecting AI assistants with Jira & Confluence for smart project management.
Git MCP Server
cyanheadsA Git MCP server for AI agents. STDIO & Streamable HTTP.
Github Mcp Server
githubGitHub's official MCP Server
GitLab MCP Server
yoda-digitalProduction-grade GitLab MCP server with 86 tools — full GitLab control from any AI agent (Claude, Cursor, Zed).
Comments