Provides tools for retrieving GitHub repository insights, including repository overview (stars, forks, license, topics, languages, last push), commit activity, stale pull requests, contributor statistics, recent releases, and issue backlog grouped by label.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@mcp-github-insights Which PRs have been stale for more than 2 weeks in sagar1708/mcp-github-insights?"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
An MCP server that exposes GitHub repository insights as tools for AI assistants — repo health, commit activity, stale PRs, contributor stats, releases, and issue backlogs.
Point any MCP client (Claude Desktop, Cursor, VS Code, Claude Code) at it and ask things like:
"Give me an overview of sagar1708/mcp-github-insights"
"Which PRs have been stale for more than 2 weeks?"
"Who are the top contributors in the last 90 days?"
Tool | What it does |
| Stars, forks, license, topics, languages, last push |
| Commits per week, top authors, latest commit (default 30d) |
| Open PRs older than N days, oldest first (default 14d) |
| Top contributors by commit count (default 90d) |
| Latest releases, newest first |
| Open issues grouped by label + oldest issues |
Related MCP server: GitHub Analytics MCP
pip install -e .Set a token for higher rate limits (optional — works without one at 60 req/hr):
export GITHUB_TOKEN=ghp_...Run the server (stdio transport):
github-insights-mcpAdd to claude_desktop_config.json:
{
"mcpServers": {
"github-insights": {
"command": "github-insights-mcp",
"env": { "GITHUB_TOKEN": "ghp_..." }
}
}
}Same shape in your MCP settings:
{
"mcpServers": {
"github-insights": {
"command": "github-insights-mcp",
"env": { "GITHUB_TOKEN": "ghp_..." }
}
}
}pip install -e ".[test]"
pytestPR review turnaround stats (median time to merge)
Code churn / hot files from commit patches
Team dashboards via scheduled digests
MIT
Commit activity for a repo over the last N days.
Returns total commits, commits bucketed per ISO week, top authors, and the latest commit.
Inputs: repo (required), owner (required), days. Returns: object
Top contributors by commit count over the last N days.
Inputs: repo (required), owner (required), days, top_n. Returns: object
Open issue backlog grouped by label, with the oldest issues surfaced.
Inputs: repo (required), owner (required). Returns: object
Most recent releases for a repo, newest first.
Inputs: repo (required), owner (required), limit. Returns: object
High-level overview of a GitHub repository.
Returns stars, forks, open issue count, license, topics, language breakdown, and how long ago it was last pushed to.
Inputs: repo (required), owner (required). Returns: object
Open pull requests that have been waiting longer than N days.
Useful for spotting review bottlenecks. Sorted oldest first.
Inputs: repo (required), owner (required), days. Returns: object
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Scored across 6 tools
Each tool targets a distinct analytical view (overview, commits, PRs, contributors, releases, issues). Minor overlap exists: commit_activity reports top authors while contributor_stats ranks contributors, and repo_overview includes issue counts while issue_backlog details issues, but descriptions clarify boundaries.
All tools use consistent snake_case noun phrases with descriptive, predictable names. No mixing of conventions or vague verbs.
6 tools is well-scoped for a read-only analytics server; each covers a distinct facet of repository insights without redundancy.
Covers core repository, commit, PR, contributor, release, and issue insights. Missing more general PR activity/review metrics and issue activity trends, but the surface is sufficient for common insight queries.
GitHub Copilot MCP: code, repo, PR, workflow tools for AI assistants.
Connect AI assistants to GitHub - manage repos, issues, PRs, and workflows through natural language.
Repo intel for AI coding agents: overview, PRs, contributors, hot files, CI, deps. Remote MCP.
Evaluates any GitHub repository for license risk, maintenance velocity, stars/forks, and security.
