- Home
- Skills
- Code Quality & Review
- Full-Stack Authority Suite
Full-Stack Authority Suite
Verify what's really live, refactor without breaking what matters, audit any "done" claim against the docs and the UI, and say it straight.
$29
Works with the AI Agents you already use
Full-Stack Authority Suite
Example session with this skill installed
Verify that my refactor of the auth logic in auth.ts is behavior-preserving. Run the existing test suite and check the login invariant.
- Read your context and instructions
- Compiled the full-stack authority suite
I have audited the refactor of auth.ts. The structural changes are contained, and all behavior-preserving gates have been verified against the baseline.
Refactor Verification Report
Invariants:
- Session tokens must be JWT-compliant.
- Failed login attempts must increment the rate-limiter in Redis.
- Unauthorized requests must return 401 Unauthorized with no body.
Before/after gates:
| Gate | Command | Baseline | Post-Refactor | Verdict |
|---|---|---|---|---|
| Unit Tests | npm test auth.spec.ts | PASS (12/12) | PASS (12/12) | PASS |
| Integration | curl -X POST /api/login | 401 (Expected) | 401 (Match) | PASS |
| Linting | eslint src/auth.ts | 0 errors | 0 errors | PASS |
Behavior comparison:
Checked JWT payload structure before and after. Both instances produced a header with alg: HS256. Rate-limiting behavior was confirmed by triggering 5 failed attempts; the 6th was correctly blocked with a 429 status code in both states.
Diff mapping:
auth.ts:42-58: Structural change to internalvalidateCredshelper. Evidence:auth.spec.tsline 102.auth.ts:110: Variable renaming for clarity. Evidence: No logic change observed in gate outputs.
Verdict: VERIFIED
Next action:
Merge the refactor branch into main. No further verification required.
Next steps
- Review the diff one final time to ensure no debug logs were left behind.
- Execute the merge command.
- Monitor the deployment pipeline to confirm the build ID matches this verified state.
Connects securely to your tools. The creator never sees your data.
About this skill
Four focused skills for agent-built software you have to trust. (1) Full-Stack Authority — verify a change layer by layer (diff, tests, build, deployment, serving, live capability) with an evidence ledger; no layer's word is taken for another's state. (2) Code Refactor Discipline — refactor with invariants stated first, baseline gates recorded before the change, the same gates after, and a claim-to-diff account of what moved. (3) Docs & UI Completion Audit — before anything is called complete, the README and shipped docs must match the functions the software actually exposes, and the capability must be exercised end-to-end in the live UI. (4) Straight-Point Protocol — turn padded, hedged status into a direct answer: verdict first, the specific uncertainty named, one concrete next step. Each skill ships with a worked example and a reference file, built from real verification work and sanitized. By DeviousDevv303.
Requires terminal access for gates. UI testing depends on accessible endpoints or local server visibility. Cannot verify third-party systems without authorization.
How to use
- 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
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 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 5 days ago