independent-review-gate
Skill2 starsChanged 4 months agoMITView on GitHub
Use when about to call complex or shippable work done, or before merging.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- mattprusak/three-wise-modelstheir own version, first seen Jul 20, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- mattprusak/three-wise-models★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository19 other files it gives its agents.
- build-release-mindsetSkill
- critical-thinkingSkill
- de-ai-proseSkill
- entry-point-auditSkill
- feasibility-guardSkill
- fix-in-the-shared-layerSkill
- fix-the-root-causeSkill
- logging-for-remote-diagnosisSkill
- no-duplicate-logicSkill
- opportunistic-fixesSkill
- performance-at-scaleSkill
- plan-with-teethSkill
- project-onto-all-systemsSkill
- question-the-premiseSkill
- tests-with-teethSkill
- user-action-edge-casesSkill
- using-instinctsSkill
- ux-designer-mindsetSkill
- verify-against-codeSkill
The file
Files
The rule
Your own review and a green test suite are not enough. You're invested in the approach, so you read past your own mistakes, and the tests only cover what you already thought to check. Before calling complex work done, have someone or something with no stake review the whole change.
A fresh set of eyes catches the defect that's obvious in hindsight and invisible to you.
This is the non-optional version of superpowers' requesting-code-review: the review is not a step to skip when the suite is green, because the green suite is exactly what can't be trusted alone — and the fix you make after the review gets re-reviewed too.
Fires when
Finishing a feature or a spec. Before merging. Before claiming "done" on anything non-trivial, or anything that ships to other people.
How to apply
Hand the full diff to an independent reviewer with explicit angles to check: correctness, edge cases, the branches no test touches, backward compatibility. Fix the blockers and cover each one with a test. Then re-review the fix itself — a second pass catches the inverted-fix mistake faster than a new test would.
The green suite is not the gate. The independent verdict is.
Worked example
Your change to a payment client passes all 200 tests and your own read-through. A reviewer with fresh eyes notices that one branch — the error path no test covers — has an inverted condition: it retries on success and gives up on failure.
The suite was green because it only ran the happy path. Your own review missed it because you "knew" what the code was meant to do, so you read what you intended instead of what you wrote.
The independent pass caught it before users did.
Red flags
| Thought | Reality |
| "Tests pass, so it's done" | Tests only cover what you thought to test. |
| "I already reviewed it myself" | You can't see your own blind spot. |
| "It's obviously correct" | The expensive bugs always look obvious until they bite. |
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.