Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →jpvarbed/gtm-diligence
Run a comprehensive pre-ship / go-to-market readiness audit of an application across security, reliability, concurrency/race conditions, accessibility (WCAG 2.2 AA), performance, and visual/UI consistency — tracing real user flows end-to-end, then returning a severity-grouped findings report, a remediation plan, quick wins, and a Safe-to-ship / Ship-with-risks / Do-not-ship recommendation.
Passed 19 of 21 test cases with the skill, and 11 of 21 without it.
and 11 more in the full report
9 fixed11 passed either way2 still failing0 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 8 test cases the skill fixed. Same model, same prompt, with and without the skill.
Following the findings section in a pre-ship diligence report for `apps/mobile-backend`, what specific structured sub-lists must be provided in the remediation deliverable?
| Checked | Without skill | With skill |
|---|---|---|
| The remediation plan categorizes required fixes into immediate low-effort improvements and longer-term architectural refactoring. | Missed | Passed |
| The action plan provides prioritized mitigation steps linked directly to the identified audit findings. | Missed | Passed |
The instructions your model reads when the skill loads.
A comprehensive, adversarial audit of a whole application before it goes to market. Unlike a diff review, this traces important flows end-to-end — codebase, architecture, data flows, API interactions, authN/authZ, state management, async operations, error handling, and user-facing surfaces — and challenges the assumptions that only hold under ideal content, a single viewport, and a cooperative user.
The output is a decision: Safe to ship / Ship with known risks / Do not ship, backed by evidence. Be adversarial in security, systematic in accessibility, precise in UI.
Credit: prompt by @shugarDadddy.
Not for: reviewing one diff/PR (use code-review), red-teaming a plan or spec (use
adversarial-review / review-council), or pure visual/3D critique (use visual-critique).
Every GTM audit confirms these first — an end-to-end audit is expensive and easy to mis-target. If the user doesn't answer, proceed only as an explicitly capped partial audit and say so.
State one, and cap the verdict accordingly (see release gate):
test cases passed, vs 11 / 21 without
More skills from the same source and nearby in the registry, with their headline benchmark lift.