MCP.so
ログイン

MCP Repo2llm Server

@crisschan

基本情報

カテゴリ

その他

ライセンス

Apache-2.0

トランスポート

stdio

作者

crisschan

投稿者

Criss Chen

MCP Repo2llm Server について

概要はまだありません

#mcp-repo2llm#code-transformation#ai-integration

MCP Repo2llm Server の設定

以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。

{
  "mcpServers": {
    "mcp-repo2llm-server": {
      "command": "uv",
      "args": [
        "run",
        "--with",
        "mcp[cli]",
        "--with-editable",
        "/mcp-repo2llm",
        "mcp",
        "run",
        "/mcp-repo2llm/mcp-repo2llm-server.py"
      ],
      "env": {
        "GITHUB_TOKEN": "your-github-token",
        "GITLAB_TOKEN": "your-gitlab-token"
      }
    }
  }
}

MCP Repo2llm Server のツール

ツール一覧はまだありません

この掲載には表示できるツール一覧がまだありません。サーバーにツールがないという意味ではありません。対応機能はプロジェクトのドキュメントをご確認ください。

ツールは README から自動的に抽出されます。メンテナーは ## Tools という見出しの下に記載することで、このタブに反映できます。

概要

What is Mcp Repo2llm Server?

Mcp Repo2llm Server is an MCP server that transforms code repositories into LLM-friendly formats, bridging the gap between traditional codebases and AI language models. It is based on the RepoToTextForLLMs project.

How to use Mcp Repo2llm Server?

Install and configure the server via uv with the required environment variables GITHUB_TOKEN and GITLAB_TOKEN. Invoke the tools get_github_repo, get_gitlab_repo, or get_local_repo with a repository URL and optional branch (default master) to receive the repository contents as formatted text.

Key features of Mcp Repo2llm Server

  • Smart Repository Scanning: processes entire codebases while preserving structure
  • Context Preservation: maintains relationships and contextual information between code files
  • Multi-language Support: language-specific optimizations for various programming languages
  • Metadata Enhancement: enriches code with relevant metadata for better LLM understanding
  • Efficient Processing: optimized for large repositories with minimal resource usage

Use cases of Mcp Repo2llm Server

  • Feed entire codebases to LLMs for analysis or code generation while retaining structural context
  • Convert GitLab, GitHub, or local repositories into a single text document for AI tool consumption
  • Standardize repository documentation and metadata for consistent LLM processing across projects
  • Enable AI-powered code review or refactoring by providing clean, structured repository text

よくある質問

1

What tokens are required to use the server?

The server requires `GITHUB_TOKEN` and `GITLAB_TOKEN` environment variables to authenticate with GitHub and GitLab respectively.

2

Which repository hosts are supported?

The server supports GitHub (`get_github_repo`), GitLab (`get_gitlab_repo`), and local file system repositories (`get_local_repo`).

3

What is the default branch used when fetching a repository?

The default branch is `master` for both GitHub and GitLab tools, but you may specify a different branch via the `branch` parameter.

4

Does the server have any runtime dependencies?

Yes, the server requires `uv` and the `mcp[cli]` package. It is run as a Python script (`mcp-repo2llm-server.py`) based on the RepoToTextForLLMs library.

5

What does the output contain?

The tools return all project information and structure from the repository as a single formatted text string, suitable for input to LLMs.

コメント

「その他」の他のコンテンツ