fix(prose-tests): what a write put in a file is evidence, not incident

[leeovery]

Summary

The symptom-gathering case failed on a claim that could not be verified — not one that was wrong. The claim: the user's ignorance was recorded in the investigation file rather than left as template placeholders. The asserter's refusal was correct:

Nothing in the prompt shows the file's post-interview content. The Edit record truncates the replacement text… the world delta names the added path only, carrying no content. Not shown, and an unshown claim does not pass.

The cap was a read's rule applied to the wrong thing. Only a read is incidental — nothing is ever claimed about bytes a walker read back out of a file, so those stay trimmed at 400. But what a write put in a file settles every claim about the artifact a phase leaves behind, and in claims mode the tool response is the only copy of that content the asserter will ever see.

The discovery cases escaped this by luck alone: their content lands via workunit create, a Bash call, which already kept 10,000 characters. Anything written by Write/Edit was invisible.

Fix: Write, Edit, NotebookEdit responses now get the produced-output ceiling alongside Bash. Reads keep the hard trim.

Test plan

• test-prose-record-action.cjs — 25/25, with two new cases: a long Edit newString survives untruncated and legible; a long Write the same
• Full proof is the case re-run above this in the stack

🤖 Generated with Claude Code

Stack

1. docs(design): prose-tests programme design log #544
2. feat(prose-tests): the framework — cases, worlds, runner, skill #545
3. test(prose): feature happy-path corpus — five worlds, seven cases #546
4. test(prose): bugfix corpus — the investigation-centric surfaces #548
5. test: retry recursive teardown removals — kill a class of phantom failures #549
6. fix(entry-skills): close the handoff fences — six files render their arms wrong #550
7. docs: a contributing page for working on the system #551
8. fix(entry-skills): every handoff arm says to invoke the skill #552
9. fix(implementation): environment setup belongs to the setup reference alone #553
10. fix(prose-tests): the asserter is told which substitutions were armed #554
11. feat(prose-tests): the mid-flow substitution, and a world only prose can describe #555
12. test(prose): claims assert consequences, not what was displayed #556
13. feat(prose-tests): record everything the agents do, results included #557
14. fix(discussion-entry): the handoff reports the source it actually had #558
15. fix(prose-tests): the stop hook records, and names the model that walked #559
16. fix(prose-tests): command output was never actually recorded #560
17. feat(prose-tests): judge the walk as told, not the summary returned #561
18. feat(prose-tests): decide in code what an agent should not be deciding #562
19. test(prose): a case starts where a session starts #563
20. feat(prose-tests): walk on Sonnet, judge on Opus, escalate a failure #564
21. test(prose): give the eight read-only cases something that can fail #565
22. test(prose): only walks that can be observed, and checks that survive the trip #566
23. fix(prose-tests): the verdict names only the model the record names #567
24. test(prose): discovery, walked to the point where work first exists #568
25. fix(prose-tests): the asserter judges which of prose or walker was at fault #569
26. docs(conventions): a step whose reference routes every exit still signposts #570
27. test(prose): discovery's epic arm, to the same durability boundary #571
28. fix(prose-tests): keep the walk's closing turn, and name the one live difference #572
29. fix(prose-tests): prescribe the cross-check, drop a check that cannot fire, lint the scope #573
30. test(prose): declare the prose each case actually walks #574
31. fix(investigation): the symptom interview never ran for work shaped in discovery #575
32. feat(prose-tests): conduct — playing a person where a script cannot reach #576
33. fix(prose-tests): what a write put in a file is evidence, not incident #577 👈 current
34. fix(investigation): the interview starts from the carrier, not over it #578
35. test(prose): quick-fix scoping, the whole definition stage in one walk #579
36. fix(prose-tests): the world log is the walker's record #580
37. fix(prose-tests): a confirmed failure reports both runs' checks #581
38. fix(prose-tests): each prescribed command runs as written #582
39. fix(scoping): context gathering reads the whole carrier, asks only the gaps #583