Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    Master-Skill: Disciplined AI Dev Loop

    1

    Disciplined end-to-end coding workflow: design gates, TDD, root-cause debugging, review, and verification.

    Free

    1 useSecurity scanned
    Master-Skill: Disciplined AI Dev Loop

    Master-Skill: Disciplined AI Dev Loop

    Example session with this skill installed

    Add a Redis-backed rate limiter middleware to our Express API (100 req/min sliding window per IP). Please implement it.

    • Read your context and instructions
    • Compiled the master-

    🧭 Router Activated: Feature Planning & TDD Phase

    Referencing references/01-brainstorming.md and references/03-tdd.md.

    1. Design & Plan (Iron Law: Design-Gate Approval)

    • Approach: Sliding-window rate limiting using Redis sorted sets (ZADD, ZREMRANGEBYSCORE, ZCARD, EXPIRE).
    • Headers: Return standard rate limit headers (RateLimit-Limit, RateLimit-Remaining, RateLimit-Reset).
    • Fallback: Fail-open with structured warning logging if Redis connection drops.

    2. Step 1: Write Failing Tests First (Iron Law 1)

    // tests/rateLimiter.test.ts
    import request from 'supertest';
    import { createApp } from '../src/app';
    
    describe('Rate Limiter Middleware', () => {
      it('allows requests within limit (under 100/min)', async () => {
        const app = createApp();
        const res = await request(app).get('/api/test');
        expect(res.status).toBe(200);
        expect(res.headers['ratelimit-remaining']).toBe('99');
      });
    
      it('rejects requests exceeding 100 requests per minute with 429', async () => {
        const app = createApp();
        for (let i = 0; i < 100; i++) {
          await request(app).get('/api/test');
        }
        const blocked = await request(app).get('/api/test');
        expect(blocked.status).toBe(429);
        expect(blocked.body.error).toMatch(/rate limit exceeded/i);
      });
    });
    

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

    About this skill

    Stop coding agents from guessing. Give them a disciplined engineering process.

    Master-Skill is a single, agent-agnostic workflow that guides any AI coding assistant through the entire software development lifecycle: design approval → planning → TDD → root-cause debugging → code review → git hygiene → proof-of-work verification.

    Instead of installing a dozen disjointed plugins, Master-Skill distills industry-standard development practices into one cohesive, zero-friction skill.


    ⚡ The Two Iron Laws (Never Skipped)

    1. No production code without a failing test first.
    2. No bug fix without root-cause investigation first.

    🧭 Built-in Phase Router

    Whenever your agent begins work, it consults the built-in phase router to load the exact reference guide needed:

    • Feature Ideation & Sizing: Classify task scope and secure human approval before touching code.
    • Task Breakdown: Break complex initiatives into bite-sized, verifiable tasks.
    • Red-Green-Refactor (TDD): Write failing test specifications first, then minimal implementation.
    • Systematic Debugging: Trace the actual mechanism of failure before proposing band-aid fixes.
    • Severity-Ranked Code Reviews: Comprehensive sanity audits for performance, security, and edge cases.
    • Git & PR Hygiene: Isolated branches, atomic commits, and clean PR descriptions.
    • Verification: Deliver indisputable proof of working functionality, not empty claims.
    • UI/UX Design Alignment: Deep integration with design specs (PRODUCT.md and DESIGN.md) before scaffolding UI.

    🔌 Universal Compatibility

    Works seamlessly across all major AI coding agents following the universal AgentSkills standard:

    • Claude Code (~/.claude/skills/)
    • Google Antigravity / Gemini CLI (~/.gemini/config/skills/)
    • Cursor / Codex / OpenCode (~/.agents/skills/)
    • Project-Local (.agents/skills/ in any repository)

    Requires the presence of the reference Markdown files cited in the router. Does not override explicit human-defined instructions in CLAUDE.md or similar project files.

    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 1 month ago

    • Free with an Agensi account

    Permissions

    Read Files
    Write Files
    Listed1 month ago
    Updated1 month ago

    What's inside

    Frequently Asked Questions

    This skill replaces "trial-and-error" coding by enforcing a structured workflow—including mandatory design gates, TDD, and root-cause analysis—ensuring your agent produces production-ready code rather than buggy guesses.

    It is fully compatible with any environment following the universal AgentSkills standard, including Claude Code, Gemini CLI (Antigravity), Cursor, and project-local configurations.

    Setup is near-instant; simply place the skill file into your agent's global or project-specific skills directory (e.g., `~/.claude/skills/`), and the agent will automatically detect and follow the new workflow protocols.

    You receive the Master-Skill core workflow file, the built-in Phase Router for different development stages, and documentation on how to integrate it with your `PRODUCT.md` and `DESIGN.md` files.

    Yes, Master-Skill is designed to be language-agnostic and works effectively across any stack, from backend microservices to complex UI/UX scaffolding guided by design specs.

    Your purchase includes lifetime access to the current version; should major updates to the AgentSkills standard or significant workflow optimizations be released, you will be notified via the marketplace dashboard.

    New in Productivity