From 3-surgeons
HARD-GATE — multi-dimensional verification that implementation meets requirements before completion
npx claudepluginhub supportersimulator/3-surgeons --plugin 3-surgeonsThis skill is limited to using the following tools:
"It works on my machine" is not verification. This gate ensures that completed work is verified across infrastructure health, regression detection, and adversarial review before anyone claims "done."
Generates design tokens/docs from CSS/Tailwind/styled-components codebases, audits visual consistency across 10 dimensions, detects AI slop in UI.
Records polished WebM UI demo videos of web apps using Playwright with cursor overlay, natural pacing, and three-phase scripting. Activates for demo, walkthrough, screen recording, or tutorial requests.
Delivers idiomatic Kotlin patterns for null safety, immutability, sealed classes, coroutines, Flows, extensions, DSL builders, and Gradle DSL. Use when writing, reviewing, refactoring, or designing Kotlin code.
"It works on my machine" is not verification. This gate ensures that completed work is verified across infrastructure health, regression detection, and adversarial review before anyone claims "done."
Do NOT claim completion until this verification passes. "Done" means verified, not "I finished typing."Invoke before ANY of these claims:
You MUST create a TodoWrite task for each step:
Run gains-gate — is infrastructure still healthy after changes? Changes can break things silently (port conflicts, stale locks, broken configs).
Sentinel re-scan on actual changes — scan the diff of what was actually implemented, not the original plan. Implementation often drifts from plan. The sentinel catches complexity introduced during implementation.
# Get the diff for sentinel input
git diff --stat HEAD~N # or appropriate range
git diff HEAD~N # full diff for sentinel content
If >3 files changed: quick consensus vote — ask all 3 surgeons: "Did we introduce regression?" with the actual diff as context. This catches issues that tests miss.
Verify tests pass — run the actual test suite. Do not claim tests pass without running them. If no tests exist for the changed code, flag this as a gap.
Counter-position: "What did we miss?" — invoke counter-position with the claim "this implementation is complete and correct." Steelman why it is not.
Present verification results — summarize all findings. Only claim completion if all checks pass. If any check fails, fix the issue and re-run verification.
The post-implementation sentinel scan differs from pre-implementation:
This catches:
Scale verification to the size of change:
| Change Size | Verification |
|---|---|
| 1 file, clear scope | Gains-gate + sentinel re-scan |
| 2-5 files | Above + test verification |
| >5 files | Above + consensus vote + counter-position |
[executing-plans] → [post-implementation-verification] → completion claim
This is the mirror of pre-implementation-review. Together they bracket the implementation phase:
pre-implementation-review → [implementation] → post-implementation-verification