# mcp-oss-onramp [Health: Active]

**Category:** 🔄 Version Control  
**Repository:** https://github.com/ryotaro-tanaka/mcp-oss-onramp.git  
**GitHub Stars:** 0  
**npm Downloads (last month):** 152  
**Views:** 1  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/mcp-oss-onramp

## Description
MCP server to scout beginner-friendly OSS issues with high precision using GitHub GraphQL

## Tools
Capabilities this server exposes over MCP:

- **scout_issues** — Scout a GitHub repository to discover truly available, beginner-friendly issues. This tool uses GitHub GraphQL for high-precision PR cross-referencing. 

IMPORTANT FOR AI AGENTS: To avoid rate limits and enable high-precision scouting, ensure the `MCP_OSS_ONRAMP_GITHUB_TOKEN` environment variable is set with a GitHub Personal Access Token. If unauthenticated, the tool fallbacks to a lower-precision REST mode.

## Claude Desktop Quick Installation
Install path detected from listing signals. Uses `npx` (confidence: high):

```json
"mcpServers": {
  "mcp-oss-onramp": {
    "command": "npx",
    "args": ["-y","mcp-oss-onramp"]
  }
}
```

## Documentation

## What mcp-oss-onramp MCP server does

The mcp-oss-onramp MCP server helps an AI agent find GitHub issues that are suitable for people beginning to contribute to open-source projects. Its selection is not limited to issues carrying an explicit beginner label. Instead, it is designed to identify beginner-friendly tasks with a focus on practical availability.

The server also checks linked pull requests. This lets it remove issues that may appear open but already have someone working on them. Issue descriptions are summarized before being returned, reducing the amount of raw text an agent must process when comparing possible contributions.

## How it works

The server uses GitHub GraphQL to inspect issue data and related pull requests. That cross-reference supports its availability check, while issue-body summarization produces more compact information for an LLM. The mcp-oss-onramp MCP server runs as a command-line MCP process, so an MCP client can start it through a command and arguments configuration.

A typical client configuration invokes `npx` with the package name and the `-y` option. Direct terminal execution is also supported. In either form, the agent can use the server to scout candidate issues instead of relying only on GitHub’s visible labels.

## Setup and configuration

Node.js 18 or newer is required. The shortest setup is:

```bash
npx mcp-oss-onramp
```

For better GitHub access, set `MCP_OSS_ONRAMP_GITHUB_TOKEN` before launching the process:

```bash
MCP_OSS_ONRAMP_GITHUB_TOKEN="your_token_here" npx mcp-oss-onramp
```

An alternative is to install the GitHub CLI and authenticate with `gh auth login`. The package can then be added to an MCP client using a server entry whose command is `npx` and whose arguments are `-y` and `mcp-oss-onramp`.

Authentication is recommended rather than mandatory. The server can run without it, but GitHub rate limits apply and the results have lower precision because linked pull requests cannot be cross-referenced.

## Tools and capabilities

The mcp-oss-onramp MCP server provides capabilities for:

- Finding GitHub issues that appear suitable for new contributors.
- Checking linked pull requests to exclude issues already being worked on.
- Detecting beginner-friendly work without requiring a beginner-specific label.
- Summarizing issue bodies for more efficient LLM consumption.
- Using GitHub GraphQL for issue and pull-request checks.

## Limitations and notes

This server depends on GitHub data and access limits. Unauthenticated execution may be less precise, both because of rate limiting and because pull-request cross-referencing is unavailable. A GitHub token or authenticated GitHub CLI session is therefore useful when the agent needs more reliable availability results.

It does not document project-specific filters, repository allowlists, or additional configuration options beyond the GitHub authentication methods and MCP process configuration described above.

_Full upstream README: https://allmcps.com/mcp/mcp-oss-onramp/readme_