Single-commit clean baseline after security scrub of niche-tells, project codenames, internal jargon, and contributor-email leaks. Contents: - 100 Rust crates (_primitives/_rust/) - 37 agent manifests (_manifests/) + generated specs (_generated/) - 67 user-invocable skills (skills/) - 33 hooks (hooks/) - Composition blocks (_blocks/) - Documentation (docs/, README.md) - TS adapter packages (_ts_packages/) - Assembler (_assembler/) - Roles (_roles/) - Templates (_templates/) - Forgejo CI (.forgejo/) Author: Denis Parfionovich <info@greendragon.info> License: see LICENSE.
1.7 KiB
1.7 KiB
Phase 4 — Output format (click)
Tell the remote agent how to shape sync-repo/sleep-results/<uuid>.md.
4a — Click
Emit ONE AskUserQuestion:
{
"questions": [
{
"question": "What format should the result use?",
"header": "Format",
"multiSelect": false,
"options": [
{
"label": "Structured markdown report",
"description": "Sections + findings + sources — default, best for research / pattern analysis"
},
{
"label": "ADR-style decision record",
"description": "Context / Decision / Consequences — best for pipeline-design output"
},
{
"label": "Checklist / action items",
"description": "`- [ ] item` list — best when you want a morning TODO"
},
{
"label": "Pros/cons table",
"description": "Markdown table with weighted criteria — best for comparative study"
}
]
}
]
}
4b — Normalise
| Label | Token |
|---|---|
| Structured markdown report | md |
| ADR-style decision record | adr |
| Checklist / action items | checklist |
| Pros/cons table | table |
Store as FORMAT.
4c — Coherence hint
Soft hint only (no re-ask), when type and format drift apart:
TASK_TYPE == compareandFORMAT != table→ hint thattableis the usual pick.TASK_TYPE == pipelineandFORMAT != adr→ hint thatadris the usual pick.TASK_TYPE == patternandFORMAT != checklist→ hint thatchecklistoften reads best.
Format: single line, does not block.
Verify-criterion
FORMAT ∈ {md, adr, checklist, table}.- Exactly ONE
AskUserQuestionin this phase.