fix(discussion-entry): the handoff reports the source it actually had

[leeovery]

Summary

• skills/workflow-discussion-entry/references/invoke-skill.md line 68 opens #### If source is \fresh` or `topic-provided`— and line 75 then statesSource: fresh`. A discussion arriving with a topic already chosen (from the epic map, or a continue menu) told the processing skill it was a direct entry.
• Now renders Source: {source}, the variable the entry skill already set.
• Found by a prose-test walker, recorded as a DEVIATION on a case that otherwise passed 5/5 — a marker is a finding in its own right, which is why they're reported even on green cases.

Scope checked before changing anything: the sibling entry skills (research, investigation) carry a Source: line only inside single-source arms, where the fixed wording is accurate. This was the one multi-source arm stating a constant.

Impact, honestly: nothing in workflow-discussion-process reads Source: — grep finds no consumer. So this is context handed downstream being wrong, not a branch being taken wrongly. Worth correcting; not worth overstating.

Test plan

• npm test green: 1699 tests, 0 fail.
• discussion-entry-seeds-from-carrier passed 5/5 with this DEVIATION recorded; re-running it should now produce the same pass with no marker.

🤖 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 👈 current
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