ReddRadar
Agent skill
audit-verify
Phase 5: Verify all corrections were applied correctly
Install this agent skill to your Project
npx add-skill https://github.com/edwinhu/workflows/tree/main/skills/bluebook-audit/skills/audit-verify
SKILL.md
Phase 5: Verify
Re-scan the corrected DOCX to confirm all fixes were applied and no new issues were introduced.
What This Phase Does
- Re-extract all footnotes from the corrected DOCX
- Re-run mechanical checks (small caps scanner, signal checker, etc.)
- Compare findings against the original audit - all flagged issues should now be clean
- Report any remaining issues
Verification Checks
- Formatting scanner: Re-run
scan_formatting.py --docx corrected.docx- expect 0 findings (or only known false positives like "Institutional Investors" in titles) - Cross-reference check: All
[_]placeholders should be resolved - Signal formatting: All signals should be italic
- Terminal periods: All footnotes should end with periods
- Id. chains: All id. references should have single-source predecessors
- Gemini re-audit on fixed footnotes: Re-run
gemini_audit.py --subset [all previously flagged FNs]to catch issues introduced by fixes or missed in the initial pass
Re-Scan Catches Real Issues
In the "Other People's Votes" audit, the Gemini re-scan on fixed footnotes caught a Wells Fargo press release title (FN206) that wasn't in the original "judgment call" list — it was in the same footnote as other fixes but wasn't flagged initially. The verify phase is not ceremonial; it finds real issues.
The verify phase MUST re-run the scanner on the corrected DOCX. Skipping verification was the root cause of missing 41 small caps fixes in the original audit.
If the re-scan finds issues, go back to Correct phase. Do NOT proceed to Archive with unresolved issues. </EXTREMELY-IMPORTANT>
Red Flags - STOP If You Catch Yourself:
| Action | Why Wrong | Do Instead |
|---|---|---|
| Skipping re-scan because "all fixes applied" | Silent failures are common | Run the scanner |
| Dismissing remaining findings as false positives | Some "false positives" are real | Investigate each one |
| Proceeding to Archive with >0 real issues | Uncorrected issues persist forever | Fix them first |
Gate: Exit Verify
Before proceeding to Archive phase:
- Re-scan completed on corrected DOCX
- Zero remaining issues (or all remaining are confirmed false positives)
- If issues found: returned to Correct phase and re-verified
Next Phase
Read ${CLAUDE_SKILL_DIR}/../../../../skills/bluebook-audit/skills/audit-archive/SKILL.md and follow its instructions.
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
edwinhu/workflows
audit-fix-loop
This skill should be used when the user asks to 'iteratively improve', 'audit and fix', 'hill-climb quality', 'grade and improve', 'score and fix', 'audit loop', 'quality loop', or needs structured iterative improvement of an artifact using scored independent audits. Also use when the user invokes a ralph loop for quality improvement rather than task completion.
edwinhu/workflows
ds-spec-reviewer
Internal skill used by ds-brainstorm at Phase 1 exit gate. Dispatches a reviewer subagent to verify SPEC.md completeness before planning. NOT user-facing.
edwinhu/workflows
pptx-render
Use when the user asks to "render pptx", "show pptx slide", "compare with pptx", "pptx to image", "export pptx slide", "original slide", "show me the original", "what does the pptx look like", or needs to extract a specific PPTX slide's content for visual comparison.
edwinhu/workflows
obsidian-organize
Organize Obsidian notes according to clawd's preferences. Use when user asks to "organize notes", "move notes to right folder", "clean up vault", "tidy vault", "file this note", or when creating new notes in the Obsidian vault. Also use when moving, renaming, or categorizing notes, or when the vault root has stray files.
edwinhu/workflows
dev-verify
This skill should be used when the user asks to 'verify completion', 'check that tests pass', 'confirm feature works', or REQUIRED Phase 7 of /dev workflow (final). Enforces fresh runtime evidence before claiming completion.
edwinhu/workflows
dev
This skill should be used when the user asks to 'start a feature', 'build a feature', 'implement a feature', 'develop', 'new feature', or needs the full 7-phase development workflow with TDD enforcement.
Didn't find tool you were looking for?