Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    Repo Health Audit

    1

    Quantitative repository auditing with an 8-dimension health score and impact-ranked remediation roadmap.

    $12

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    Repo Health Audit

    Repo Health Audit

    Example session with this skill installed

    Run a full health audit on the local repo at ./projects/payment-gateway. Include a history secret scan and provide a ranked list of fixes to improve the score.

    • Read your context and instructions
    • Compiled the repo health audit

    Audit complete for repository payment-gateway. The mechanical scan and manual judgment pass are finalized.

    Repo Health Audit: payment-gateway

    Overall Score: 68/100 (Grade: C+)
    Source LOC: 42,500
    Task Capacity: 124 mineable commits

    Buyer-Fit Signals

    • Architecture: PASS - Logic is separated from infrastructure.
    • Test Coverage: WARN - Core engine has 80% coverage, but edge cases are untested.
    • Security: FAIL - 3 potential secrets found in historical commits.
    • Documentation: PASS - README and API docs are current.
    • Maintenance: PASS - Consistent commit activity over last 6 months.

    Health Dimensions

    • Architecture (8/10): Strong layering in /src/core, but /src/api has tight coupling.
    • Test Coverage (6/10): High coverage in utilities; low coverage in transaction workflows.
    • Security Hygiene (4/10): Detected hits in config/legacy_settings.py history.
    • Docs & Onboarding (9/10): Excellent docstrings and setup guides.
    • Code Cleanliness (7/10): Moderate cyclomatic complexity in auth_provider.py.
    • History & Maintenance (8/10): Clear commit messages and distinct author identities.
    • Dependency Health (7/10): 4 packages are 2+ major versions behind.
    • CI/CD Maturity (9/10): Fully automated pipeline with linting and unit tests.

    Priority Fix List

    Security: Purge historical secrets from config/legacy_settings.py. Use BFG Repo-Cleaner or git filter-repo. (Impact: +15 pts)
    2.

    Test Coverage: Add integration tests for src/transactions/processor.py to cover partial failure states. (Impact: +8 pts)
    3. Architecture: Decouple src/api/routes.py from the database models by introducing DTOs. (Impact: +5 pts)
    4. Dependency: Update express and jsonwebtoken to current stable versions. (Impact: +4 pts)

    Next steps

    • Review metrics.json in the workdir for the raw hit counts.
    • Run the suggested git filter-repo command to clear security warnings.
    • Schedule a follow-up audit after the transaction test suite is expanded.

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

    About this skill

    Technical debt and architectural decay often go unnoticed until a repository becomes unmaintainable or fails a due diligence check. This package provides a structured audit system to quantify code health, identify security risks, and generate an impact-ranked roadmap for remediation.

    What it does

    • Multi-dimension auditing evaluates architecture, test coverage, security, documentation, code cleanliness, maintenance history, dependency health, and CI/CD maturity.
    • Automated scanning uses a Python-based engine to collect objective metrics including LOC, commit history, and secret detection.
    • Standardized scoring applies a consistent rubric to produce a 0-100 health grade, ensuring results are comparable across different timeframes or projects.
    • Security hygiene identifies potential secrets in git history without exposing the sensitive values in logs or reports.
    • Actionable fix lists generates a prioritized list of tasks ordered by their potential impact on the overall health score.

    How it works

    1. Repository Access: Point the skill to a local directory or provide a GitHub URL for cloning.
    2. Mechanical Scan: The skill executes audit_repo.py to generate a metrics.json file containing raw data points.
    3. Rubric Application: Metrics are measured against the provided scoring rubric and buyer-fit signals.
    4. Judgment Pass: The agent performs manual code inspection for qualitative factors like layering and error handling.
    5. Report Delivery: A Markdown report is generated containing the score, dimension breakdown, and prioritized fix list.

    Frameworks & tools

    The system utilizes a Python-based scanner, Git for history analysis, and a structured Markdown rubric. It integrates with GitHub for repository metadata and cloning.

    Why this beats prompting it yourself

    Generic prompts often result in vague "this code looks clean" feedback. This skill enforces a rigid scoring rubric and uses a mechanical scanner to back up qualitative judgments with hard data like commit volume and dependency counts.

    Use cases

    • Pre-acquisition audit to evaluate the technical health and risk of a codebase before a purchase or merger.
    • Refactoring roadmap to identify which files or modules are the biggest bottlenecks to development velocity.
    • CI/CD integration to track health score trends over time and catch architectural drift.

    Known limitations

    History secret scanning can take several minutes on large repositories. The skill operates in a read-only mode and will not automatically push fixes to the repository.

    History secret scanning is time-intensive on large repos. Operates read-only; does not automatically commit fixes. Requires Python3 to run the included scanner script.

    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

    Python

    Permissions

    Terminal / Shell
    Browser
    Read Files
    Write Files
    Network Access
    Environment Variables
    Listed9 days ago

    What's inside

    Frequently Asked Questions

    The audit supports any language for architectural and history analysis, though the scanning engine is optimized for deep dependency and complexity metrics in Python, JavaScript, and TypeScript environments.

    No, the skill operates in a strictly read-only capacity to ensure your codebase remains untouched; it provides the prioritized roadmap, but code changes must be applied by your developers or a separate coding agent.

    You receive the core Python-based scanning engine, the standardized scoring rubric, and the prompt templates required to generate the comprehensive Markdown health reports.

    While the mechanical scan is fast, the 'Judgment Pass' and secret detection can take 5–10 minutes for large repositories with extensive git histories or thousands of files.

    This skill is designed for any LLM-based agent with file system access or GitHub integration, such as AutoGPT, BabyAGI, or custom LangChain implementations.

    The skill identifies the presence of secrets (like API keys or passwords) within your git history without logging the actual values, allowing you to rotate them safely without leaking them to the LLM.

    New in Agents & Orchestration