Works with the AI Agents you already use
Claude Code
Cursor
Codex CLI
Muse
Dots+21 moreClaude CodeCursorCodex CLIMuseDots
Manus
Grok
Grok Bot+18 agentsReview Panel: Devil's Advocate Multi-Round QA Gate
Multi-round QA gate with independent reviewer roles, deduplicated findings and a clear PASS/BLOCK verdict per round.
$5
Review Panel: Devil's Advocate Multi-Round QA Gate
Example session with this skill installed
Set up a two-round review panel for my marketplace listing, then tell me whether round 2 can ship.
- Read your context and instructions
- Compiled the review panel
What happened
The skill generated the role cards for a listing review, then evaluated the gate for round 2 in three situations: an open P1, no files for the round, and a clean round. Real output from the bundled scripts:
$ python scripts/panel_plan.py --type listing --rounds 2
# Round 1
## Role: Buyer
**Mandate:** Read this listing as a potential buyer deciding whether to purchase. ...
**End your response with this exact line:** COUNTS P0=n P1=n P2=n
## Role: Red Team ... (8 role cards, each with a COUNTS line)
$ python scripts/findings_ledger.py gate --dir review --round 2
GATE BLOCK — round 2: P0=0 P1=1 open (exit 1)
$ python scripts/findings_ledger.py gate --dir only-round-1 --round 2
GATE BLOCK — no findings files for round 2 (exit 1)
$ python scripts/findings_ledger.py gate --dir clean --round 2
GATE PASS — round 2: P0=0 P1=0 (exit 0)
Next steps
- Fix the open P1 and run the reviewers for round 2 again.
- Save each reviewer's findings with the round number in the file name.
- Ship only when the gate reports PASS for the round you name.
Connects securely to your tools. The creator never sees your data.
About this skill
Use when a deliverable needs a structured multi-round review with fixed reviewer roles (buyer, red team, fact check, format) before shipping. The skill generates role cards that each end with a COUNTS line, merges and deduplicates findings across rounds, and evaluates a gate for the round you name: PASS only when that round has findings files and zero open P0 and P1.
Do not use when you are reviewing a code diff line by line; use a code review skill for that. Do not use when a single reviewer is enough.
Requires: Python 3.9 or later (standard library only).
Limits: Does not guarantee that the panel finds every error or that the deliverable is correct after passing. The gate counts open P0 and P1 from the COUNTS lines only and does not read external systems.
Does not guarantee that the panel finds every error or that the deliverable is correct after passing. The gate counts open P0 and P1 from the COUNTS lines only and does not read external systems.
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 8 days ago
Requires
Python