fix(investigation): the interview starts from the carrier, not over it

[leeovery]

Summary

#575 made the symptom interview run for work shaped in discovery. The first walk through the arm showed what it does when it finally executes: three of six questions re-asked things the carrier had already answered and initialisation had already written into the file — asking what the user sees when it crashes, against a Symptoms section already reading "No error surfaces to the user — the page just fails."

That is what the prose literally said to do. The Otherwise arm loaded the question bank with no instruction to read the file it had just seeded, and the question bank carries no such instruction either. The don't-re-ask rule lives in the entry skill ("Do not re-ask; live conversation context, when present, supplements the carrier"); the process skill lost it at the handoff.

Pre-existing, newly reachable. The arm was effectively unreachable for carrier-shaped work until #575 fixed the guard, so the gap has existed as long as the arm has — it simply never ran on the path that matters.

Fix — three sentences ahead of the Load: read what the Symptoms section already holds (the seed is the user's own account), ask what it does not answer, confirm rather than re-ask where it is thin.

Provenance

Found by investigation-gathers-symptoms (#576) on its first run — the walker put Q1 to a user whose file already contained the answer. One walk did this, the confirmation walk did not, which is exactly the model-dependent behaviour an unstated intent produces.

Test plan

• Conventions lint 29/29; prose gate 90/90
• Verified by the case re-run once the stack is reloaded — its further claim "nothing the user already said in discovery is put back to them as a question" is the direct probe

🤖 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
34. fix(investigation): the interview starts from the carrier, not over it #578 👈 current
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