feat(prose-tests): conduct — playing a person where a script cannot reach

[leeovery]

Summary

• conduct is the third per-case input, beside answers and stubs: a description of the person on the other side, not a queue of replies. A scripted answer covers a question with one right response; symptom gathering explores until it judges it has enough, so there's no fixed number of turns to script. The walker answers as the described person would, in their words, for as long as the prose keeps asking. A case may carry both — the script covers discrete gates, the description covers open stretches.
• Two guards keep it from becoming a back door through P4. The template states it describes the user and never the walk — it cannot name an arm, end a step, or say what the prose ought to do. And the walker's standing instruction now says playing the user is not steering the walk: a described user who would happily stop talking is still not permission to cut a loop the prose has not ended.
• First case to demand it ships with it: investigation-gathers-symptoms walks investigation through its symptom interview. Its claim 3 — the interview runs despite the carrier having populated Symptoms — is the claim that fails on the prose before fix(investigation): the symptom interview never ran for work shaped in discovery #575's guard fix, so the finding lands twice as CLAUDE.md requires. Its user relays the bug rather than living it: concrete about what they saw, plain about what they don't know, volunteering nothing.
• Corpus at 12 cases; validation covers the new field (present-but-empty rejected).

First run results

The case ran and did its job — 6/6 path steps including claim 3, all four deterministic checks, world PASS. It FAILed on one further claim whose evidence the harness truncates (an Edit response cut at 400 chars) — that's the next PR up the stack. It also surfaced the re-asking prose gap — the PR above that.

Test plan

• Prose gate 90/90; conventions lint 29/29; corpus valid
• Full verification is the re-run after the two fixes above land (needs a plugin reload for the walker-definition change)

🤖 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 👈 current
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
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