Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    GitHub Intelligence

    1

    Deterministic GitHub repo health scores (A-F), maintenance-risk detection, and side-by-side repo comparisons for any AI agent.

    Free

    1 useSecurity scanned
    GitHub Intelligence

    GitHub Intelligence

    Example session with this skill installed

    Analyze the health of the facebook/react repository and check for any maintenance risks.

    • Read your context and instructions
    • Compiled the github intelligence

    I have completed a health analysis for facebook/react. The report is based on live metrics from the GitHub API.

    Repository Health Report: facebook/react

    Verdict: Grade A. This is a highly active, well-maintained repository that is safe for production adoption.

    MetricValue
    Overall Score94/100
    Letter GradeA
    Stars225,430
    Open Issues1,240
    Last Commit2024-05-22
    Contributors1,650+

    Component Breakdown

    • Activity (25%): 98/100 (High commit volume and active author count)
    • Issue Health (20%): 88/100 (Efficient triage of new reports)
    • PR Health (20%): 92/100 (Low volume of stale pull requests)
    • Contributor Health (15%): 100/100 (Very high contributor count, low bus factor)
    • Release Activity (10%): 95/100 (Consistent release cadence)
    • Documentation (10%): 100/100 (Comprehensive documentation and license detected)

    Detected Risks

    [Medium] Stale Issues: There are over 400 issues older than 90 days, though this is expected for a project of this scale.

    [Low] Contributor Concentration: No single contributor dominates the commit history, indicating a healthy distributed team.

    Next steps

    1. View the full visual dashboard: https://github-intelligence-mcp-ashy.vercel.app/repo/facebook/react
    2. Compare this against vuejs/core if you are evaluating frontend frameworks.
    3. Review open pull requests to see current development priorities.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Gives any AI agent deterministic GitHub repository intelligence: 0–100 health scores with A–F grades, maintenance-risk detection (stale issues/PRs, bus factor, release gaps), and side-by-side repo comparisons. Pairs with the open-source GitHub Intelligence MCP server (10 read-only tools + resources + prompts).

    Unlike asking an LLM to guess repo metrics from stale training data, every number here is computed in code from the live GitHub REST API — fast, cheap, and reproducible.

    What your agent can do:

    • Analyze any public repo: health score, letter grade, 6 weighted components with evidence
    • Detect maintenance risks: stale issues/PRs, inactivity, bus factor, contributor concentration, release gaps — each severity-tagged
    • Compare two repos side by side before adopting a dependency
    • List issues, PRs, commits, contributors, releases — all structured JSON

    Use it for: dependency due diligence, triaging a project before contributing, auditing maintainer health, choosing between libraries, portfolio/repo review.

    Live dashboard: https://github-intelligence-mcp-ashy.vercel.app/ Source: https://github.com/Dayyan-Saeed/Github-Intelligence-Mcp

    Read-only access. Cannot create issues or PRs. Requires a GitHub fine-grained PAT and the GitHub Intelligence MCP server installed. API rate limits apply.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    1 use

    Times used to date

    Free forever

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 26 days ago

    • Free with an Agensi account

    Requires

    GitHub
    Github Intelligence MCP
    Github Intelligence Ashy MCP
    Git

    Permissions

    Read Files
    Write Files
    Network Access

    Allowed Hosts

    api.github.com
    Listed26 days ago

    What's inside

    Frequently Asked Questions

    Unlike standard LLMs that rely on outdated training data, this skill fetches real-time metrics directly from the GitHub REST API to ensure your repository health scores and 'bus factor' assessments are accurate as of this second.

    You will need a GitHub Personal Access Token (PAT) and a compatible MCP (Model Context Protocol) host, such as Claude Desktop or any AI agent that supports MCP servers.

    No, this is a read-only auditing tool designed for risk analysis and due diligence; it does not have the permissions to modify code, post comments, or manage issues in your repositories.

    The skill uses a deterministic, weighted algorithm that evaluates six key dimensions—including issue resolution speed, contributor diversity, and release cadence—to generate a 0-100 score and a letter grade.

    Yes, the 'Library Comparison' feature allows your agent to pull data for two different repositories simultaneously, providing a side-by-side risk and activity audit to help you choose the safer dependency.

    The skill is designed to work with any public repository or private repository that your GitHub Personal Access Token has permission to view.

    New in LLM Ops