repolens

    1

    Read-only GitHub repository inspector for tracking PR staleness, file churn, issue health, and CI status.

    $1

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    repolens

    Example session with this skill installed

    Check octocat/Hello-World for any stale pull requests over 14 days and tell me if the main branch is passing CI.

    • Read your context and instructions
    • Compiled the repolens

    I have checked the status of octocat/Hello-World. Here is the current health report based on live repository data.

    Repository Health Audit: octocat/Hello-World

    MetricStatus / Details
    CI Status (main)✅ Success - Latest run #442 completed successfully 3 hours ago.
    Stale PRs (>14d)2 PRs identified as inactive.
    Oldest Stale PR#12 "Refactor parser" by @carollee (80 days inactive)
    Recent Activity5 commits in the last 7 days.
    Open Issues14 open (excluding PRs).

    Stale Pull Request Details

    • #12: Refactor parser
      • Author: carollee
      • Last Updated: 80 days ago
      • Status: Needs review or update; high inactivity.
    • #11: Fix typo in docs
      • Author: bobbuilder
      • Last Updated: 19 days ago
      • Status: Minor update, likely forgotten.

    Next steps

    • Ping @carollee on PR #12 to check if the refactor is still relevant.
    • Review the diff for PR #11 to quickly merge the documentation fix.
    • Run get_file_churn if you want to see which files these PRs are touching most frequently.

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

    About this skill

    Inspect GitHub repositories with real-time data through a read-only MCP server interface. This skill eliminates guesswork by fetching live repository states, including issue backlogs, PR diffs, commit histories, and CI status. It provides deep visibility into repository health without the risk of accidental mutations.

    What it does

    • Analyze PR diffs by fetching full unified diffs and per-file change summaries.
    • Identify stale PRs by calculating days since the last update against a custom threshold.
    • Track file churn to pinpoint the most frequently modified and potentially risky files.
    • Monitor CI status by retrieving recent GitHub Actions workflow runs and conclusions.
    • Search code across the repository using GitHub syntax to find definitions or usages.

    How it works

    1. Supply the repository in owner/name format to establish a read-only connection.
    2. Request specific data using commands like get_file_churn or get_pr_diff.
    3. Review the synthesized report, which prioritizes actionable data like stale counts or failing builds.
    4. Dive deeper into specific files or commits using targeted searches or history limits.

    Frameworks & tools

    This skill interacts with the GitHub API via the RepoLens MCP server. It handles ISO date filtering, unified diff parsing, and GitHub-scoped code search syntax.

    Why this beats prompting it yourself

    Standard LLMs rely on training data that is months or years old, making them useless for checking current PR status or CI failures. This skill bridges the gap by pulling live, structured data directly from GitHub's API into the context window.

    Use cases

    • Audit repo health by checking CI status and flagging ignored pull requests.
    • Perform technical due diligence by identifying high-churn files and risky code areas.
    • Prepare for releases by filtering open issues for blockers and checking build conclusions.
    • Review complex PRs by summarizing large diffs and checking cross-file impacts.

    Known limitations

    Read-only access; cannot create issues, merge PRs, or push code. Large issue backlogs are truncated at 100 items. Code search requires a valid GitHub token for reliability.

    Read-only access only. Large repos may experience 100-item truncation on issues. Rate limits apply to unauthenticated requests (60/hr) and code searches (10/min).

    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

    Fresh listing

    Recently published to Agensi

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed 9 days ago

    Requires

    GitHub
    Python
    Git

    Permissions

    Terminal / Shell
    Read Files
    Write Files
    Network Access
    Environment Variables

    Allowed Hosts

    github.com
    api.github.com
    Listed9 days ago

    What's inside

    Frequently Asked Questions

    No, repolens operates through a strictly read-only interface. It is designed for inspection and analysis—such as tracking PR staleness and CI status—without the risk of accidental code mutations or repository changes.

    This skill requires a GitHub Personal Access Token (PAT) to authenticate with the API. Once configured, you simply provide the repository in the 'owner/repo' format to begin fetching live data.

    While standard LLMs rely on outdated training data, repolens fetches real-time data directly from the GitHub API. This allows it to report on the exact current state of your CI builds, the latest commit history, and active PR diffs that happened minutes ago.

    Repolens is built on the Model Context Protocol (MCP), making it compatible with any AI agent or platform that supports MCP servers, such as Claude Desktop or various IDE-integrated AI tools.

    To maintain performance and stay within context window limits, the skill truncates large issue backlogs at 100 items and provides per-file change summaries for very large diffs. This ensures the most relevant health metrics are prioritized without overwhelming the agent.

    New in LLM Ops