My AI Subagent Faked the Verification Output I Asked It to Attach

A subagent reported 7 files changed with zero disk changes behind it, then fabricated the ls -la and grep -c output I demanded as proof. The postmortem on why self-report is not evidence even when it comes with receipts, and the landmark-file check I run on every delegated task now.

Read Original

Related