Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    API Architect (Free Starter)

    1

    A 12-point rigorous audit for API endpoints to catch permanent design mistakes before they ship.

    Free

    4 usesSecurity scanned
    API Architect (Free Starter)

    API Architect (Free Starter)

    Example session with this skill installed

    Review this endpoint: POST /v1/orders. It takes a customer_id and a list of items. Returns 200 on success. No pagination or idempotency keys mentioned in the draft.

    • Read your context and instructions
    • Compiled the api architect free

    API REVIEW: POST /v1/orders, verdict: redesign
    Fails: 6 (no pagination), 7 (missing idempotency), 3 (failure cases unspecified), 11 (no examples)
    Flags: 8 (timestamps not defined)
    The one fix that matters most: Define idempotency keys to prevent duplicate orders.

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

    About this skill

    The problem

    API mistakes become permanent the moment a consumer depends on them. Most developers rely on vibes during code reviews rather than a rigorous standard for naming, error handling, and future-proofing.

    What it does

    • Evaluates endpoints against 12 critical design questions covering idempotency, pagination, and naming.
    • Identifies specific failures in error response shapes and status codes.
    • Verifies if types, units, and timezones are explicitly defined rather than implied.
    • Provides a structured verdict card with a single "must-fix" priority and a clear ship/fix/redesign status.
    • Automatically degrades to a core 4-point check when only a route name is provided.

    Why this beats prompting it yourself

    This skill enforces evidence-based reviews over intent-based assumptions. It prevents "rubber-stamping" by requiring specific evidence for every pass and triggers a redesign verdict automatically if more than three core failures are detected.

    Use cases

    • Run a final gate check on a pull request before merging API changes.
    • Audit existing legacy endpoints to document technical debt and breaking changes.
    • Validate new API designs during the RFC or contract-drafting phase.
    • Standardize API review quality across a distributed engineering team.

    Known limitations

    Requires a design, contract, or diff for a full 12-point audit. If only a route name is provided, it can only perform a partial assessment.

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Code Quality & Review.

    Requires a schema or diff for a full 12-point audit. With only a route name, it performs a partial 4-point review.

    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

    4 uses

    Times used to date

    Free to use

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 2 months ago

    • Free with an Agensi account

    Permissions

    No special permissions declared or detected

    Listed3 months ago
    Updated4 days ago

    What's inside

    Frequently Asked Questions

    This skill uses a rigorous 12-point audit framework to identify permanent design flaws in naming, error handling, and idempotency before you ship, moving beyond subjective 'vibes-based' code reviews.

    For a full 12-point audit, provide an API design document, an OpenAPI/Swagger contract, or a code diff; if you only provide a route name, the skill automatically scales down to a 4-point core check.

    Yes, this skill is designed to work across any AI agent platform that supports text-based skills, making it ideal for integration into your existing PR review or RFC workflows.

    You receive the full logic for the 12-point auditing framework, the automated verdict card system, and the 'must-fix' prioritization engine to standardize reviews across your team.

    Absolutely. It is highly effective for auditing legacy endpoints to document technical debt or identifying necessary breaking changes for versioning updates.

    If the audit detects more than three core failures, the skill automatically triggers a 'Redesign' verdict, preventing the common mistake of rubber-stamping fundamentally flawed API contracts.

    New in Code Quality & Review

    Free