GSD Verifier Agent
The agent that doesn't trust summaries - it reads the actual code, checks the actual wiring, and answers the only question that matters: did we achieve the goal or just complete the tasks?
The GSD Verifier agent specification - covering goal-backward verification methodology, the critical distinction between task completion and goal achievement, three-level artifact verification (existence, substance, wiring), must-have derivation from phase goals, re-verification mode for targeted gap checking, and the VERIFICATION.md output format that drives fix-verify loops.
Also connected to
The read-only auditor that answers one question: does what was built match what was planned - and if not, is the deviation justified or is it scope creep?
The living strategy document behind ClaudeOps - a 5-stage constitutional governance model where Claude Code thinks differently at each development stage, with explicit quality gates, error-type differentiated remediation, and the pilot project data that validated the entire approach in 55 minutes instead of 7.5 hours
Concrete implementation planning
The implementation plan for a ClaudeOps dashboard that gives visual oversight of every project's stage progression - portfolio view, gate status queue, blocker aggregation, and velocity metrics, built on the existing Vite/React/FastAPI stack rather than starting from scratch
Living document on process infrastructure for AI development