I made Evidence Graph, every SKILL instruction 100% enforced, in 20 languages

github.com/wrtnlabs/evidence I made an open source program which can enforce AI agents to keep every skill instructions. Every rule, requirement, schema, and API becomes an obligation the check enforces. 100% coverage of every requirement. 100% compliance with every principle. 20+ languages, plus Markdown and Swagger. Modern AI already writes code, documents, and fiction well. The remaining bottleneck is not creation but adherence: an agent can understand a specification, ignore one constraint, and still declare the work complete (The Compliance Gap). Evidence makes the agent account for every instruction where the work is done, stating how the output satisfies it or why it does not apply. That single obligation turns compliance from a promise into a complete, reviewable graph. typescript /** @evidence docs/discount.md#coupon-stacking States the per-issuer stacking limit this section defines, in the buyer's words. @evidence POST:/orders/{orderId}/coupons Explains the rejection this endpoint returns for an over-stacked coupon set. @evidence ../hooks/useCouponStacking.ts#useCouponStacking Renders the limit this hook resolves. @evidence .agents/skills/principles/SKILL.md#no-hard-coding Renders limits from props instead of branching on known issuer names. @evidenceExclude .agents/skills/principles/SKILL.md#fix-root-causes No failure path exists in a pure renderer. / export function CouponStackingNotice(props: IProps): JSX.Element