fix(implementation): environment setup belongs to the setup reference alone

[leeovery]

Summary

The entry skill asked for setup instructions and wrote the document the processing skill also writes — while its own header said "information gathering only. Do NOT execute any setup commands at this stage. The processing skill contains instructions for handling environment setup." Whichever ran first won and the other silently took its already-exists arm. A prose-test walker hit exactly that on both runs.

One file owns it now. workflow-implementation-process/references/environment-setup.md checks whether the document exists, asks when it does not, and stores the answer. By implementation time it exists and is ready to use; if it somehow does not, the same reference creates it. No special handling, no second owner.

So the entry sheds it entirely:

• Step 4: Check Environment removed (Invoke becomes Step 4)
• references/environment-check.md deleted
• the handoff drops its Environment: line, which existed only to carry an answer the entry no longer gathers
• environment-setup.md returns to exactly what it was on main

An earlier version of this PR split ownership — entry asks, process records — which only relocated the confusion. This replaces it.

The implementation-picks-first-task prose case is updated to the settled design: the entry hands off without touching the environment, and the setup step consumes the answer and records the document.

Test plan

• npm test green: 1699 tests, 0 fail. Corpus valid, 13 cases.
• environment-setup.md is byte-identical to main — verified.

🤖 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 👈 current
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