Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    repo onboarding doc

    1

    Generate high-density developer onboarding guides by scanning code manifests, CI configs, and architectural patterns.

    Free

    Security scanned
    repo onboarding doc

    repo onboarding doc

    Example session with this skill installed

    Scan this repository and generate an onboarding guide. A new backend engineer is joining today and needs to get the API running locally and understand the project structure.

    • Read your context and instructions
    • Compiled the repo onboarding doc

    I have scanned the repository, analyzed the package.json and .github/workflows, and traced the request flow from src/index.ts. I found two stale setup steps in the README which I have corrected below.

    Onboarding: Payment-Gateway-Service

    What this is

    This is a Node.js microservice for processing cross-border payments. It acts as a wrapper around the Stripe and Adyen APIs, providing a unified internal interface for the frontend checkout flow.

    Architecture in 60 seconds

    The service follows a layered architecture: src/api/routes.ts (Entry) -> src/services/ (Business Logic) -> src/adapters/ (Third-party APIs).

    graph LR
      Client -> Router[src/api/routes.ts]
      Router -> Service[src/services/payment.service.ts]
      Service -> Adapter[src/adapters/stripe.adapter.ts]
      Service -> DB[(Postgres)]
    

    Run it locally

    1. Copy the env template: cp .env.example .env
    2. Start the infrastructure: docker-compose up -d
    3. Install dependencies: npm install
    4. Run migrations: npm run db:migrate
    5. Start the dev server: npm run dev
      You should see "Server listening on port 3000" and a successful DB connection log.

    Repo tour

    PathWhat lives hereWhen you'll touch it
    src/api/Express routes and middlewareAdding new endpoints or auth logic
    src/services/Core payment processing logicChanging how transactions are calculated
    src/adapters/Stripe/Adyen integration codeUpdating third-party API versions
    src/models/TypeORM entitiesChanging the database schema
    tests/integration/Jest integration testsVerifying full payment flows

    Everyday tasks cookbook

    Add a new payment provider

    1. Create a new class in src/adapters/ implementing IPaymentProvider.
    2. Register the provider in src/services/provider.factory.ts.

    Run specific tests

    1. Run npm test tests/services/payment.service.spec.ts for unit tests.

    Gotchas

    • Webhook Callbacks: Local testing requires stripe-cli to forward webhooks to port 3000.
    • Node Version: The project is pinned to v18.16.0 in .nvmrc; higher versions cause bcrypt build errors.
    • DB Seeding: npm run db:seed must be run after migrations or the user roles will be missing.

    Open questions for the team

    • TODO: Confirm if the REDIS_URL in .env.example is still required as no files currently import a redis client.

    Next steps

    • Review the src/adapters/ folder to see how existing providers are implemented.
    • Check the Open questions section with the tech lead during the sync.

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

    About this skill

    Onboarding new developers to a legacy or complex codebase is a slow, manual process. This skill automates the repo tour by scanning entry points, manifests, and CI configurations to build a high-density orientation guide. It ignores stale README claims and looks at the source of truth—the code and build scripts—to identify how data flows from router to database.

    What it does

    • Architectural mapping traces the request path through routers, controllers, and services to visualize the data flow.
    • Dependency analysis extracts project metadata and external service requirements from manifests like package.json or go.mod.
    • Command extraction identifies the actual build, test, and migration scripts used in CI/CD rather than relying on outdated docs.
    • Gotcha harvesting locates technical debt clusters, TODO comments, and local environment quirks that trip up new hires.
    • Repo touring generates a directory map explaining exactly which folders to touch for specific feature updates.

    How it works

    1. Inventory phase scans root manifests, env templates, and CI configs to verify the project's real-world setup.
    2. Architecture trace follows entry points to map how business logic is distributed across the codebase.
    3. Validation step flags discrepancies between what the README claims and what the code actually does.
    4. Document generation outputs a structured guide including a local setup cookbook and a list of verified technical gotchas.

    Frameworks & tools

    Works with any language using standard manifests including Node.js, Python, Go, Rust, Java, and .NET. It analyzes Docker, Terraform, and GitHub/GitLab CI configurations.

    Why this beats prompting it yourself

    Generic prompts often hallucinate setup steps or repeat outdated info from the README. This skill cross-references manifests with CI scripts and actual file paths to ensure every instruction is grounded in the current codebase.

    Use cases

    • Orient a new engineer on their first day with a copy-pasteable setup guide.
    • Document a legacy repository that has zero existing documentation.
    • Prepare for a hand-off between teams by mapping core business logic locations.
    • Audit a project to find undocumented environment variables and setup gaps.

    Known limitations

    Large repositories may result in a trunk-only analysis where deep sub-directories are summarized rather than detailed. It requires read-access to manifests and CI files to be effective.

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for DevOps & Cloud.

    Requires manifest files (package.json, pyproject.toml, etc.) and CI configs to be present. Large repos focus on entry points and top-level directories.

    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 buying a skill.

    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

    Free to use

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 23 days ago

    • Free with an Agensi account

    Permissions

    Read Files
    Network Access
    Listed23 days ago

    What's inside

    Frequently Asked Questions

    Unlike standard AI prompts that summarize existing (and often outdated) README files, this skill cross-references your actual source code, CI/CD pipelines, and configuration manifests to generate a 'source of truth' guide based on how the app actually runs.

    The skill is language-agnostic and supports any project with standard manifests, including Node.js, Python, Go, Rust, Java, and .NET, as well as infrastructure tools like Docker and Terraform.

    This skill identifies 'gotchas' by scanning for technical debt clusters, undocumented environment variables in templates, and inconsistencies between CI build scripts and local setup instructions.

    You receive the full repo onboarding skill logic, ready to be integrated into your AI agent or development workflow, along with documentation on how to provide the necessary file access for analysis.

    For extremely large monorepos, the skill focuses on a high-level 'trunk-only' analysis to ensure accuracy, providing summaries for deep sub-directories rather than granular detail for every single file.

    New in DevOps & Cloud

    Free