plan

Workflow Plan

Read workflow/product.md and generate objectives + tasks matched to worker capability. Advances phase to plan.

Do NOT generate tasks if workflow/product.md is missing or phase is past 'plan'. Warn and stop.

Step 1: Check prerequisites

python3 -c "
from workflow_kit.runtime.state import load_state
from pathlib import Path
s = load_state(Path('.'))
print('phase:', s.phase)
print('product_type:', s.product_type)
"
ls workflow/product.md 2>/dev/null && echo "product-ok" || echo "product-missing"
ls workflow/worker_capability.json 2>/dev/null && echo "capability-ok" || echo "no-capability-warning"

• Phase execute or later: "Already past planning. Use /workflow-kit:execute to continue."
• workflow/product.md missing: "Run /workflow-kit:init first."
• No capability file: warn but continue — orchestrator will plan conservatively.

Step 2: Generate workplan

python -m workflow_kit workplan

Orchestrator (DeepSeek) reads:

• workflow/product.md (Vision/Mission/Core)
• workflow/worker_capability.json (skill scores)
• workflow/tasks/completed/ (what's already done)

Creates 3–5 objectives aligned to Vision. Tasks under each objective matched to worker skill scores. Tasks needing a skill < 3.0 are decomposed into smaller sub-tasks.

Step 3: Write workflow/workplan.md

# Workplan — <product name>

Generated: <timestamp>

## Objectives

### <Objective 1>
**Why:** aligned to Vision/Mission
**Success criteria:** measurable outcome

Tasks:
- [ ] <task-id>: <description>

Step 4: Advance state to 'plan'

from workflow_kit.runtime.state import load_state, save_state
from pathlib import Path
s = load_state(Path('.'))
if s.phase == "define":
    s.transition_to("plan")
    save_state(s, Path('.'))
    print("Phase: define → plan")

Step 5: Report

Workplan generated:

  Objectives: N
  Tasks:      N pending in workflow/tasks/pending/

  obj-01: <name>  (N tasks)
  obj-02: <name>  (N tasks)

Review workflow/workplan.md, then:
  /workflow-kit:execute  — start the dispatcher with reviewer agents