Verify consequential AI work
Ten local-first checks for consequential AI work. Trace where data moved, review what an agent can remember, compare AI drafts with accepted work, right-size agent access, diagnose failed runs, and verify claimed outcomes against durable systems of record. More
Skills 10 in the ChatGPT listing
Agent Autopsy Reconstruct a completed, failed, expensive, or confusing agent session to find where its trajectory broke and what one change would prevent recurrence. Bottleneck Shift Compare a workflow before and after AI assistance to determine whether work disappeared or moved into review, waiting, correction, exception handling, or rework. Did It Land Reconcile an AI agent's completion claims with durable evidence in the relevant system of record. Human Review Tax Compare an AI-produced artifact with the accepted final artifact and quantify the human correction and review it required. Policy Vs Reality Compare written AI policy with observed agent configuration and runtime evidence, producing a non-accusatory mismatch ledger. Right Size My Agent Compare an agent's configured capabilities with observed tool use and propose a least-privilege configuration for defined workflows. Safe to Paste Inspect text or files before they are shared with an AI service, make a task-preserving redacted copy, and explain residual exposure. Skill Sunset Decide whether an installed agent skill should be kept, narrowed, revised, merged, quarantined, or retired using real activation and outcome evidence. What Does My AI Remember Audit persistent information an AI agent can reuse across sessions, including provenance, sensitivity, staleness, contradictions, scope, and deletion evidence. Where Did My Data Go Reconstruct the data journey of a specific AI task from local traces, tool configuration, logs, and artifacts. 
