From meta-methodology-write-verification
Write verification protocol - never report success without verifying work was actually saved. Re-read files after edits, verify changes exist, only report success after verification passes.
How this skill is triggered — by the user, by Claude, or both
Slash command
/meta-methodology-write-verification:meta-methodology-write-verificationThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
> **Quick Guide:** Never report success without verifying work was actually saved. Re-read files after edits, verify changes exist, only report success after verification passes.
Quick Guide: Never report success without verifying work was actually saved. Re-read files after edits, verify changes exist, only report success after verification passes.
<critical_requirements>
(Never report success without verifying your work was actually saved)
(Re-read files after completing edits)
(Verify changes exist in the file before reporting success)
(If verification fails, re-attempt the edit - do NOT report success)
</critical_requirements>
<write_verification_protocol>
CRITICAL: Never report success without verifying your work was actually saved.
Agents can:
This causes downstream failures that are hard to debug because the agent reported success.
After completing ANY file edits, you MUST:
Include this in your final validation:
**Write Verification:**
- [ ] Re-read file(s) after completing edits
- [ ] Verified expected changes exist in file
- [ ] Only reporting success after verification passed
</write_verification_protocol>
<agent_types>
For code changes (frontend-developer, backend-developer, tester):
For documentation changes (documentor, pm):
For structural changes (skill-summoner, agent-summoner):
For configuration changes:
</agent_types>
<critical_reminders>
(NEVER report task completion based on what you planned to do)
(ALWAYS verify files were actually modified before reporting success)
(A task is not complete until verification confirms the changes exist)
(Re-read files after edits to confirm changes were saved)
(If verification fails, re-attempt the operation)
</critical_reminders>
npx claudepluginhub agents-inc/skills --plugin meta-methodology-write-verificationEnforces validation rules for Claude Code tools (Edit, Write, Bash, NotebookEdit) to prevent errors like unread files before edits, non-unique old_strings, and missing directories. Useful before file changes or after failures.
Audits post-implementation plans against changes by parsing checkboxes for completeness and accuracy. Supports autonomy modes (Autopilot, Critical, Verbose) and file-review. Invoke via /verify-plan.
Spawns a separate verifier agent to inspect diffs against acceptance criteria after a worker returns SUCCESS, catching plausible-but-wrong code, partial implementations, and scope drift before commit.