verify-before-done
Skill5 starsChanged 46 days agoMITView on GitHub
use this when about to claim work is complete, fixed, or passing — before commit, PR, or “landed+verified” — run fresh verification and read the evidence first.
11 other projects use a version of this.
Adopted by
Most recently taken up by
- fvkmizel/soulkeepertheir own version, first seen Nov 8, 2026
- reddygopalkrishna/developer-agent-skillstheir own version, first seen Nov 8, 2026
- Hvftpie13/Flashmind_Gemini-3.8-Flash_Smart-agenttheir own version, first seen Nov 1, 2026
- KMaruthi2002/universal-llm-skillstheir own version, first seen Oct 25, 2026
- joaoguirunas/team-ostheir own version, first seen Oct 21, 2026
- jsvillalbat/sevito-skillstheir own version, first seen Oct 20, 2026
Other versions11 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 11
- Hvftpie13/Flashmind_Gemini-3.8-Flash_Smart-agent★ 39same name
- reddygopalkrishna/developer-agent-skills★ 8same name
- joaoguirunas/team-os★ 6same name
- GusevV1987/ai-workspace-kit★ 2same name
- MPIsaac-Per/claude-code-loop-patterns★ 2same name
- rrrrrredy/beforedone★ 2same name
- yuchi-chang/no-cape★ 2same name
- KMaruthi2002/universal-llm-skills★ 1same name
- fvkmizel/soulkeeper★ 0same name
- jsvillalbat/sevito-skills★ 0same name
1 more not listed
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 repository49 other files it gives its agents.
- AGENTS.mdAGENTS.md
- maintainers/AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- buying-researcherSkill
- cursor-cloud-agents-whenSkill
- debug-anthropicSkill
- debug-pocockSkill
- debugSkill
- discover-the-ideaSkill
- docs-google-styleSkill
- golang-safetySkill
- golang-securitySkill
- golang-testingSkill
- grok-acpSkill
- host-config-pyinfraSkill
- lang-cppSkill
- lang-csharpSkill
- lang-cSkill
- lang-dartSkill
- lang-dockerSkill
- lang-goSkill
- lang-javaSkill
- lang-js-tsSkill
- lang-kotlinSkill
- lang-luaSkill
- lang-makefileSkill
- lang-phpSkill
- lang-powershellSkill
- lang-protobufSkill
- lang-pythonSkill
- lang-rubySkill
- lang-rustSkill
- lang-shellSkill
- lang-sqlSkill
- lang-swiftSkill
- lang-terraformSkill
- language-routerSkill
- lang-web-markupSkill
- modern-pythonSkill
- pr-reviewSkill
- sdlc-artifactsSkill
- sdlc-onboardingSkill
- security-hardeningSkill
- shell-safetySkill
- tddSkill
- tracker-sdlcSkill
- ui-craftSkill
- ux-designSkill
- yagniSkill
The file
We split this skill into the rules it gives an agent
4 rules. 0 are common, 4 are distinctive, and 0 are very specific to this project.
0
Common
Most of the 10 edited versions we read keep these.
4
Distinctive
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 20
If you have not run the proving command in this turn, you cannot claim pass, fixed, green, or done.
- Distinctive: line 24
Before any success claim, commit, PR, or “next task”:
- Distinctive: line 43
Operator in the loop: run verification for evidence — bots oversee workers.
- Distinctive: line 73
Stop. Run the proof. Or say what is still unverified.
“Common” is measured: kept by at least 6 in 10 of the 10 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Verify before done
Rewrite of obra/superpowers verification-before-completion @ b36e0829. Compress, not a paste. Id: verify-before-done.
Notify only when the work is landed+verified: see SDLC Definition of done.
The proving command is always re-run. Verifier mint, resume, and never the builder: see SDLC Item agents.
Iron law
No completion claims without fresh verification evidence.
If you have not run the proving command in this turn, you cannot claim pass, fixed, green, or done.
Gate
Before any success claim, commit, PR, or “next task”:
- Identify the command (or check) that proves the claim.
- Run it fresh and complete — not a remembered earlier run.
- Read full output, exit code, failure count.
- Match output to the claim. If not, report actual status with evidence.
- Only then claim success — and cite the evidence.
Skip a step = claiming, not verifying.
Always / Ask first / Never
| Always | Ask first | Never |
| Fresh command for this claim | Softening verify for throwaway spikes you will delete | “Should pass” / “probably” / “seems” |
| Full suite or scoped command that actually covers the claim | Partial checks when full suite is expensive — name what you skipped | Trusting agent “success” without VCS/diff or independent run |
| Red-green proof for new regression tests | Operator in the loop when evidence is ambiguous or flaky | Satisfaction (“Great!”, “Done!”) before evidence |
| Line-by-line checklist for requirements DoD | Unsupervised push to protected / prod deploy / secret rotate | Extrapolating from linter to build, or build to product fix |
Operator in the loop: run verification for evidence — bots oversee workers. If security or CI marks a check flaky/MEDIUM, do not self-clear — show the operator the output and wait.
Claim → evidence
| Claim | Requires | Not enough |
| Tests pass | Test run: 0 failures | Prior run, “should pass” |
| Linter clean | Linter: 0 errors | Partial path |
| Build succeeds | Build exit 0 | Linter green |
| Bug fixed | Symptom repro now passes | Code changed |
| Regression test works | Fail without fix, pass with fix | Passes once |
| Agent finished | Diff on project-main or trunk (the land target) shows the change | Agent said success; item branch only |
| Requirements met | Checklist against plan/ticket | Tests alone |
Roles
| Role | Owns | Does not own |
| builder | Running this gate before ship claims | Skipping security on changes to skill homes, reviewed at Review |
| tester | CI that produces evidence; verifier subagent for the item | Claiming product DoD from fmt-only green |
| security | Intake / security clear at Review | Day-to-day verify coaching |
| manager | After-act landed+verified | Blessing without evidence |
Red flags
- Wording that implies success without a command in this turn
- “Just this once” / “I’m tired” / “agent said it’s fine”
- Moving on because the diff “looks right”
- Calling the item landed when it only exists on an item branch
Stop. Run the proof. Or say what is still unverified.
Upstream pin
Intent pin: obra/superpowers verification-before-completion @ b36e0829c6d0140e93cfef2ca599b1b07d4a7797. See SOURCES.md.
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.