What problem does it solve? Teams need a fast, honest verdict on whether a code change is ready to ship, without wading through long findings lists or trusting a green CI badge that never exercised the changed behavior. ## Core Features & Use Cases - A-F Lane Grading: Scores correctness, security and permissions, data and state, domain truth, UX and release behavior, tests and verification, and deploy readiness, then combines them into one overall grade. - Instant-F Triggers and Grade Caps: Automatically fails changes with hardcoded secrets, injection, auth bypass, payment flaws, or destructive migrations, and caps grades on auth, payment, migration, and public-API surfaces without specific evidence. - Ship Gate Verdict: Maps the overall grade to a ship, ship-with-caveats, or hold recommendation with required upgrades named for any grade below B. - Use Case: Before merging a pull request that touches a payment webhook, ask for a grade; the skill checks Instant-F triggers, verifies webhook signature handling and error paths, applies the payment-surface cap, and returns a letter grade with the exact fixes needed to raise it. ## Quick Start Ask the agent to use suede-code-grader on the current diff or pull request and return the overall grade, required upgrades, and ship gate verdict.
