What problem does it solve? AI coding agents produce more code than anyone reviews, and without a written quality bar they quietly lower standards — adding @ts-ignore suppressions, skipping tests, or weakening thresholds to get checks passing. This Skill turns your project's quality bar into a written, mechanically checkable contract that outlives any single session. ## Core Features & Use Cases - Guided constraint interview: Asks four questions with sane defaults, detects your stack from package.json, pyproject.toml, and CI config, then writes a CONSTRAINTS.md with numbered thresholds for coverage, security, performance, and accessibility. - Tool-backed enforcement: Maps each dimension to a de facto tool (tsc, Semgrep, gitleaks, osv-scanner, Lighthouse, axe-core, Stryker) with install commands, run commands, and lifecycle placement from edit loop to CI. - Bar-weakening guards: Watches diffs for lowered thresholds, silenced checkers, skipped tests, stubs, and unauthorized exceptions, with a reference floor-guard implementation. - Use Case: Before running an autonomous build loop on a TypeScript web app, use this Skill to record an 80% changed-line coverage gate, zero high-severity dependency vulnerabilities, and LCP under 2500ms, so the agent cannot ship code that fails your standards. ## Quick Start Ask the agent to set up project constraints and quality gates for this repository, then let it interview you and write CONSTRAINTS.md.
