audit(token_audit): one submission carrying two slash commands is one message

[EdbertChan]

/reflect /cat-mode on a single line injects the same user text twice, tens
of milliseconds apart. The verbatim-repeat rule read that as a re-send, the
strongest single frustration signal, and escalated a pass to FAIL on it.

A repeat now needs at least five seconds between the two rows. A genuine
re-send minutes later still counts, asserted both ways.

Co-Authored-By: Claude Opus 5 (1M context) noreply@anthropic.com
Claude-Session: https://claude.ai/code/session_01KU2pPKob4MJ1NqjsfTNyYJ

Depends-On: #509

---

Note

Low Risk
Narrow heuristic tweak in offline transcript auditing with unit tests; may slightly reduce verbatim-repeat detection for duplicates sent 1–4 seconds apart.

Overview
Fixes a false positive in reflect’s verbatim-repeat frustration signal when one user submission produces two identical transcript rows milliseconds apart (e.g. /reflect /cat-mode on one line).

frustration_signals in token_audit.py now requires at least MIN_RESEND_GAP_SECS (5) between matching normalized user texts before counting a verbatim-repeat (still capped at 10 minutes). Sub-second duplicates are ignored; a real re-send minutes later still flags and can still drive intervention-must-automate.

Two Codex unit tests lock in the near-duplicate vs. minutes-later behavior.

^{Reviewed by Cursor Bugbot for commit 8c65d80. Bugbot is set up for automated code reviews on this repo. Configure here.}

[EdbertChan]

This pull request is part of a Mergify stack:

#	Pull Request	Link
1	hook: a script handed to the user is a claim that it runs	#506
2	audit(token_audit): count "bogus" and "it didn't run" as the same class	#507
3	hook: refuse a quoted command string passed through a login shell	#508
4	gate: a branch taken per operating system needs a test that injects one	#509
5	audit(token_audit): one submission carrying two slash commands is one message	#510	👈
6	audit: use the transcript's own markers for what the human actually sent	#511

[cursor]

Bugbot couldn't run - usage limit reached

Bugbot is counted against Cursor usage for this user or team, and this run hit a usage or spend limit.

A user or team admin can review and increase usage limits in the Cursor dashboard.

(requestId: serverGenReqId_d9473ccb-6cba-4d85-af59-5cfaf02e0463)