Stats
Actions
Tags
Help us improve
Share bugs, ideas, or general feedback.
From metacognitive-guard
Enforces verification checklist before claiming completion on code changes, bugs, or claims. Checks builds/tests/output; outputs markdown report with confidence and readiness.
npx claudepluginhub ancplua/ancplua-claude-plugins --plugin metacognitive-guardHow this command is triggered — by the user, by Claude, or both
Slash command
/metacognitive-guard:verification-before-completionThe summary Claude sees in its command listing — used to decide when to auto-load this command
# Verification Before Completion > **If you didn't run it, you don't know if it works.** ## Triggers Activate when about to say: "done", "complete", "finished", "works", "fixed", "the implementation is ready", "this should work". ## Verification Checklists ### Code Changes - [ ] Build passes (no errors) - [ ] Tests pass (or explicit reason why skipped) - [ ] No new warnings introduced - [ ] Actually ran the code and showed output ### Factual Claims - [ ] Verified against assertions.yaml - [ ] WebSearch if claim involves dates/versions/status - [ ] Source cited ### Bug Fixes - [ ] ...
/fire-double-checkVerifies work completion claims by running npm build, test, typecheck, and lint checks on current work or specified target, reporting results with evidence and exit codes.
/verifyVerifies if current work is complete, fixed, or passing before committing by invoking the jig:verify skill.
/verifyVerifies code changes before completion claims by auto-detecting stack, running tests/lints/type-checks/builds, and providing checklists with actual outputs.
/verify-fixVerifies code fixes work by running tests (vitest/pytest/go), checking UI/API via curl/Playwright, or builds/configs; reports success/failure with evidence.
/verifyVerifies branch code via tests, lint (auto-fix option), type-check, visual review, and simplification. Produces phase statuses (passed/failed/skipped), overall status, and evidence paths.
/verifyVerifies code, architecture, or bug fixes using Software Engineering Chain of Verification (SE-CoVe). Supports --quick, --thorough, and --focus flags for depth and focus areas like security or performance.
Share bugs, ideas, or general feedback.
If you didn't run it, you don't know if it works.
Activate when about to say: "done", "complete", "finished", "works", "fixed", "the implementation is ready", "this should work".
## Verification Checklist
### Build Status
- [ ] Build - [PASS/FAIL]
- [ ] Tests - [PASS/FAIL/SKIPPED: reason]
### Steps Performed
1. [What you did]
2. [What output you saw]
### Confidence
[high/medium/low] - [why]
### Ready to Claim Complete
[YES/NO] - [if NO, what's missing]
"This should work" → actually test it. "The fix is complete" → run the build. "I've implemented the feature" → show test output.