
Verification Before Completion
Makes the agent run real checks and show evidence before saying work is done or passing
- Repository
- obra/superpowers
- Stars
- 310,931
- Licence
- MIT
- Files
- 1 file
What it does
This skill makes the agent stop and run the actual verification command (such as tests, a linter, or a build) and read the real output before it says a task is finished, fixed, or passing. It also covers checking an agent's own success reports against the code changes. The skill is a set of rules and checklists written in plain instructions, so it changes how the agent behaves rather than adding a new tool.
Great for
- Developers who have seen an AI agent claim tests pass when they were never run
- Teams that want a verification step before commits or pull requests
- People delegating coding work to sub-agents who want the results checked against the actual diff
Skip it if
- Projects with no test, lint, or build commands, since there is nothing to run as evidence (the skill cannot create verification on its own)
- Non-technical tasks such as writing or design work where there is no command that proves success
- Quick throwaway edits where the extra verification step adds delay without much benefit
Try asking
- “Before you commit, check that the tests really pass and show me the output”
- “Fix this bug and confirm the original failing test now passes before you tell me it's fixed”
- “Did the sub-agent actually make the changes it reported? Check the diff before we merge”
- “Make sure the build succeeds, not just the linter, before we open the pull request”
Install in 3 steps
- 1
Download the skill
The ZIP contains the 1 original files, the MIT licence and an attribution note.
verification-before-completion.zip - 2
Add it to your agent
Claude Code (all projects)
mkdir -p ~/.claude/skills && unzip verification-before-completion.zip -d ~/.claude/skills/Claude Code / Agent SDK (this project only)
mkdir -p .claude/skills && unzip verification-before-completion.zip -d .claude/skills/Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.
- 3
Ask for the task
No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.
Get better results
- Name the exact verification command you want run (for example npm test or cargo build) so the agent does not guess
- Ask for the regression test red-green cycle when fixing bugs, so the test is shown to fail before the fix and pass after
- Give the agent the original plan or requirements so it can build a line-by-line checklist to verify against
What's inside
- SKILL.md
Redistributed unmodified from obra/superpowers under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.