IMPLEMENTAUDIT
On this page
  1. Verified complete
  2. Audit gap repair loop
  3. Blocked handoff
  4. Unsafe or unbounded stop
  5. How to inspect it
  6. End-state matrix
  7. Completion anti-patterns
First run

Possible endings

A run should not end just because the chat stopped.
It ends by proving completion, handing off a real blocker, or refusing unsafe or unbounded work.
Audit gaps are repair loops, not final endings.

What to look for

AUDIT_COMPLETE must appear before IMPLEMENTAUDIT_RUN_COMPLETE.
If it cannot, the run should leave a handoff or stop record with the blocker and next owner action named.

Verified complete

The final audit found final, checkable evidence for every in-scope item.
The run may then emit AUDIT_COMPLETE, followed by IMPLEMENTAUDIT_RUN_COMPLETE.

AUDIT_VERIFY checks and deliverables rechecked
AUDIT_COMPLETE every in-scope item has final checkable evidence
IMPLEMENTAUDIT_RUN_COMPLETE valid only after audit completion

Audit gap repair loop

The final audit found a gap, but the run had a bounded fix path.
It records the gap, applies the fix, reruns the relevant evidence, and repeats the final audit loop.

AUDIT_GAPS missing or weak evidence found
countermeasure chosen with a changed approach or new evidence
AUDIT_VERIFY rerun before closure

Blocked handoff

The run cannot honestly finish without an owner decision, missing credential, unavailable environment, or scope decision.
It should name the blocker and the smallest next owner action.

AUDIT_HANDOFF not complete
blocker: owner decision required
next owner action: approve route or provide missing evidence

Unsafe or unbounded stop

If the request asks for:

  • unauthorized mutation
  • release, publication, or provenance
  • destructive filesystem behavior
  • an unbounded scope jump

the run stops instead of inventing authority.

Important: install, commit, push, tag, release, publish, and provenance remain separate gates.
No ending on this page grants those actions by implication.

How to inspect it

Work recordFor compact work, locate the bound work record through the final response, final audit ledger, cited artifacts, and marker order. For phased work, also open .IMPLEMENTAUDIT/runs/<task-slug>-<id>/.
State fileFor phased work, check STATE.md; for compact work, check the final audit ledger, cited artifacts, and transcript markers.
Final audit ledgerLook for closure, gap, or handoff evidence.
Transcript markersCheck marker order: audit completion before run completion.

End-state matrix

EndingWhat you should seeWhat it means
Verified completeAUDIT_COMPLETE before IMPLEMENTAUDIT_RUN_COMPLETEIn-scope items have final checkable evidence.
Audit gap repair loopAUDIT_GAPS, new evidence, then audit againThe final check found a gap and looped back through repair before any ending.
Blocked handoffAUDIT_HANDOFF or ANDON_HANDOFFA real blocker remains and the next owner action is named.
Unsafe stopSTOP or handoff textThe request is unbounded, unauthorized, or unsafe to mutate.

Completion anti-patterns

  • A stopped chat is not valid closure.
  • A passing-looking rerun without cause analysis is not valid closure.
  • A hidden substitution is not valid closure.
  • A completion marker after a handoff is not valid closure.