verified-gtm-skills
claudeGTM plays as agent-loadable skills. Three orthogonal labels per play instead of one word doing three jobs: provenance (verified-source / standard-practice / original), gating (executable — T3 eval fixtures and T1 shape lints executed in CI — or documented, for T2 asserts and T0 rubrics that nothing executes), and hash-pinned attestations that expire. No play claims to be proven in real work.
23 skills
0