draft-plans

name	draft-plans
description	Graph-node skill: reads .ship-manifest.json (or story/epic IDs), launches one background agent per story to write plans/*.md files, applies critique checklist, updates DB with plan_file paths. Artifact-chain-aware version of /draft-plan. Consumes manifest from /plan-stories, produces plans/*.md for /env-preflight.
triggers	["/draft-plans","/draft-plans .ship-manifest.json","/draft-plans story-NNN story-NNN","/draft-plans epic-NNN"]
args	[{"name":"args","type":"string","description":"Manifest path (.json), story IDs (story-NNN), epic IDs (epic-NNN), --briefing <path>, and/or --skip-critique. At least one input required.\n"}]

Draft Plans Skill Invoked

User has requested: /draft-plans {{args}}

Phases

Read each phase file as you enter it.

1. Resolve → Read resolve.md

   • Parse args, detect input mode (manifest/ID/mixed)
   • Resolve stories, classify frontend/fast-path

2. Prepare → Read prepare.md

   • Read critique checklist, generate plan names
   • Decision lookup, exemplar matching, preference prediction

3. Write plans → Read plan-writer.md

   • Frontend gemini_design calls
   • Fast-path: write inline. Agent-path: launch background agents
   • Collect results, frontend validation

4. Critique → Read critique-loop.md

   • Skip if --skip-critique or caller handles separately
   • Delegate to subagent: 5-lens self-critique, Gemini escalation, contract gate
   • Report results

Child files

• resolve.md — Parse args, resolve stories, frontend/fast-path detection
• prepare.md — Critique checklist, plan names, decisions, exemplars
• plan-writer.md — Plan templates, agent prompts, result collection
• critique-loop.md — Critique subagent, Gemini escalation, report