From ultrapowers
This skill should be used when about to claim work is complete, fixed, or passing — before committing, creating PRs, or telling the user 'done'. Requires running verification commands and reading output before making success claims.
npx claudepluginhub jaidhyani/jai-cc-plugins --plugin ultrapowersThis skill uses the workspace's default tool permissions.
Evidence before claims, always.
Verifies tests pass on completed feature branch, presents options to merge locally, create GitHub PR, keep as-is or discard; executes choice and cleans up worktree.
Guides root cause investigation for bugs, test failures, unexpected behavior, performance issues, and build failures before proposing fixes.
Writes implementation plans from specs for multi-step tasks, mapping files and breaking into TDD bite-sized steps before coding.
Evidence before claims, always.
No completion claims without fresh verification evidence in the current response. "Should work now" is not evidence. Run the command, read the output, then state the result.
This applies before ANY transition: completing a task, moving to the next task, delegating to an agent, committing, creating a PR, or telling the user "done." Skip any step and the claim is unverified.
Before any claim of success, completion, or readiness:
| Claim | Requires | Not Sufficient |
|---|---|---|
| "Tests pass" | Test output showing 0 failures | Previous run, "should pass" |
| "Build succeeds" | Build command exit 0 | Linter passing |
| "Bug fixed" | Failing test now passes | "Code changed, should work" |
| "Regression test works" | Pass → revert fix → fail → restore → pass | Test passes once |
| "Requirements met" | Line-by-line check against spec | "Tests pass" |
| "Agent completed work" | Review the actual diff | Agent's self-report |
About to write any of these? Stop and run the verification first:
Unverified claims break trust, ship broken code, and waste time on rework. "I'm confident" is not evidence. Run the command.