c43e931bdb
The serialization think step (step 2) had a hardcoded list of generic task types (research_plus, draft, review, polish, etc.) — these NEVER matched CLP templates. This is the root cause of auto-hired agents and quick-task fallbacks for all book/story/article pipelines. Replaced with: - Instruction to use ONLY task types from TEAM ROSTER supported_templates - Explicit sequencing rules for book, article, blog, recipe, short story pipelines - Book pipeline: book_research first, cascade handles the rest - Article pipeline: ai_article_research first - Blog: blog_research per post - Recipe: recipe_collection_plan - Short story: short_story Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
105 lines
4.8 KiB
YAML
105 lines
4.8 KiB
YAML
name: planning
|
||
description: "Rotating chair deliberation → structured work breakdown → child tasks created."
|
||
debug: true
|
||
system: agent_prompt
|
||
participant_prompt:
|
||
- "= identity.md"
|
||
agent_prompt:
|
||
- "= identity.md"
|
||
sections:
|
||
- agent
|
||
- project
|
||
- history
|
||
- rag
|
||
- participants
|
||
- participants_prompt
|
||
- roster
|
||
- message
|
||
- instructions
|
||
steps:
|
||
- type: think
|
||
rotate_participants: true
|
||
loop:
|
||
max_iterations: 3
|
||
hint: |
|
||
You are {agent.name}. This is round {task.iteration} of the planning session.
|
||
The group is breaking down this project or task into a concrete work plan.
|
||
Write YOUR perspective on the plan from your area of expertise:
|
||
- What work streams or tasks need to exist?
|
||
- What order do they need to happen in? What depends on what?
|
||
- Which agents from the TEAM ROSTER are best suited to each piece?
|
||
- What risks, gaps, or open questions do you see?
|
||
- What do you agree or disagree with from prior planning rounds above?
|
||
Reference the TEAM ROSTER above for agent names and the task types they own.
|
||
Use EXACT task type names from each agent's supported_templates list — these are the
|
||
template filenames the Bouncer uses to route tasks. Do NOT invent task type names.
|
||
Be specific about agent assignments — use EXACT names from the TEAM ROSTER.
|
||
ROUNDTABLE CHAIR RULE: For business strategy or operations deliberations, assign the chair
|
||
to a business-register agent (director or executive role). Do NOT assign pure writing or
|
||
creative-execution agents as chair for planning deliverables.
|
||
When the group has genuinely agreed on a complete plan, include exactly:
|
||
"consensus_reached: true"
|
||
If the plan still has gaps or unresolved assignments, do NOT include that line.
|
||
- type: think
|
||
agent: first_available
|
||
hint: |
|
||
Structure the team's agreed plan into a precise task list.
|
||
IMPORTANT: Do NOT re-deliberate or reconsider what was agreed above.
|
||
Your only job in this step is to serialize the decisions into clean task records.
|
||
For each task:
|
||
- task_name: short, specific
|
||
- description: full detail — enough that the assigned agent can execute without asking
|
||
- agents: EXACT name(s) from TEAM ROSTER
|
||
- task_type: MUST be an EXACT template name from the agent's supported_templates in the TEAM ROSTER
|
||
- priority: 1 (critical) to 5 (nice-to-have)
|
||
Order tasks correctly — blockers before dependents.
|
||
No ambiguous assignments. No placeholder names.
|
||
Do NOT invent task type names. Each task_type MUST appear verbatim in the TEAM ROSTER.
|
||
|
||
BOOK/FICTION SEQUENCING RULE:
|
||
- For novels and books, the pipeline is: book_research → book_outline → book_chapter (×N) →
|
||
chapter_review (×3 per chapter) → chapter_roundtable → chapter_polish → book_editorial → project_index
|
||
- Do NOT spawn book_chapter tasks until book_outline has run and provided chapter context.
|
||
- Start with ONE book_research task only. The cascade templates handle everything after that.
|
||
|
||
ARTICLE SERIES SEQUENCING RULE:
|
||
- For article series: ai_article_research → ai_article_plan → ai_article_write (×N)
|
||
- Start with ONE ai_article_research task only.
|
||
|
||
BLOG SEQUENCING RULE:
|
||
- For a blog post: blog_research → blog_write (one per post)
|
||
|
||
RECIPE SEQUENCING RULE:
|
||
- For a recipe collection: ONE recipe_collection_plan task only.
|
||
|
||
SHORT STORY RULE:
|
||
- For a short story (under 20k words): ONE short_story task only.
|
||
|
||
CANONICAL SELECTION RULE: If the plan produces multiple competing drafts of the
|
||
same artifact (e.g. three mission statement drafts, two roadmap versions), the plan
|
||
MUST include a final "Select Canonical: [artifact]" task assigned to the most
|
||
relevant specialist that explicitly picks one version, merges the best elements, and
|
||
archives the rest. Do NOT leave competing versions unresolved.
|
||
|
||
DECOMPOSITION RULE: When the deliverable is a batch of similar items (e.g. 10 podcast
|
||
scripts, 5 chapters, 8 articles), create ONE individual task per item — NOT a
|
||
single "write all 10" batch task. Each item must be its own task with its own specific
|
||
description. Review and polish tasks can be batched at the end.
|
||
|
||
- type: package
|
||
packet_type: PlanningResponsePacket
|
||
schema:
|
||
tasks:
|
||
- task_name: "string — short descriptive name"
|
||
description: "string — full instructions for the assigned agent"
|
||
agents:
|
||
- "string — EXACT agent name from TEAM ROSTER"
|
||
task_type: "string — MUST be a template name from AVAILABLE TASK TEMPLATES"
|
||
priority: "integer 1 (critical) to 5 (nice-to-have)"
|
||
insert_children: true
|
||
|
||
- type: close
|
||
rag_update: false
|
||
adjudication:
|
||
enabled: false
|