What problem does it solve?
It prevents teams from falsely concluding that work is “done” by verifying whether the intended outcome is actually achieved using concrete evidence.
Core Features & Use Cases
- Goal-backward verification: Restates the desired outcome as a concrete, testable statement before checking anything.
- Evidence-based checks: Confirms what must be true, what must exist (with content), and what must be connected (integrations, wiring, and test execution).
- Risk surfacing: Identifies where the outcome will likely break via edge cases, error paths, missing validation, and configuration gaps.
- Use Case: Before closing an issue or merging a PR, verify that the user-visible behavior matches the goal—not that the checklist was merely completed.
Quick Start
Ask your AI to produce a verification report using the “state the goal, then verify TRUE/EXISTS/CONNECTED, and summarize PASS/PARTIAL/FAIL with line-cited evidence” output format for your change.