did-it-actually
Verify Claude completed the requested work with a contract, fresh-context critic, and capped re-audit loop.
Verify Claude completed the requested work with a contract, fresh-context critic, and capped re-audit loop.
did-it-actually decomposes a user request into falsifiable acceptance criteria, records implementation claims, and audits the result with deterministic checks plus a fresh-context critic. Missing or unsupported claims trigger a fix-and-re-audit loop capped at three iterations, with a mechanically grounded VERIFIED or NOT VERIFIED verdict. It checks request fidelity rather than serving as a code reviewer, security scanner, or test framework.
Resource types
Use cases
Platforms
Capabilities
Diagnose and repair Claude Desktop third-party model configuration gaps
Turn an existing website into a deployable conversational agentfront.
Public GitHub facts last synced Jul 10, 2026.
Turn a repository's structure into an editable, interactive architecture map.
Install coding guidance distilled from s4.codes videos as a Claude skill.