Multi-agent council using real subagent spawning (Task() tool), deterministic CLI state management, and file-based handoffs. Orchestrator stays lean while specialists execute in fresh contexts.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
A direct command skips the review prompt. Inspect the source before running it.
Multi-agent council using real subagent spawning (Task() tool), deterministic CLI state management, and file-based handoffs. Orchestrator stays lean while specialists execute in fresh contexts.
Multi-Agent Council โ Agent Team Coordination
Production-grade multi-agent coordination using real subagent spawning via Task(), deterministic CLI for all structural operations, and file-based handoffs between agents. Each agent runs in its own fresh 200k context. The orchestrator stays lean (~10-15% context usage).
Iron Law
Every council operation that touches file naming, numbering, state transitions,
or validation MUST go through the planning-tools CLI. Agents execute โ
the CLI manages structure.
This is non-negotiable. LLMs cannot reliably handle sequential numbering, state machine transitions, or structural validation. The CLI does this deterministically. Agents focus on what LLMs are good at: reasoning, writing, and analysis.
When to Use
Activate council coordination when:
Task spans 3+ files or 2+ systems
Task requires research before implementation
Task benefits from independent specialist perspectives
Context would exceed 50% for a single session
User requests: "start a council", "use team mode", "team session"
Do NOT use when:
Task is a single-file change with clear requirements
Task is a quick bug fix with obvious cause
User explicitly wants a single-session approach
Anti-Shortcut Rules
These are hard constraints. No exceptions without explicit user override.
#
Rule
Why
1
YOU CANNOT skip CLI commands and create council files manually
CLI enforces naming, numbering, and format. Manual creation causes state drift.
2
YOU CANNOT role-switch in a single session instead of spawning agents
Context burns fast, quality degrades, no isolation between perspectives.
3
YOU CANNOT advance state without a validated handoff
Unvalidated handoffs propagate errors to downstream agents.
4
YOU CANNOT skip quality gates (use --force only with user permission)
Gates catch 80% of handoff quality issues before they compound.
5
YOU CANNOT let agents communicate directly
All handoffs go through files. This is the ONLY reliable state transfer.
6
YOU CANNOT manually edit BOARD.md
Board is generated output. CLI regenerates it from state.
Common Rationalizations
Thought
Reality
"I can just role-play all agents in one session"
Context burns fast, quality degrades, no parallelism, no isolation
"File handoffs are unnecessary overhead"
They're the ONLY reliable state transfer between fresh contexts
"The CLI is unnecessary ceremony"
LLMs cannot reliably number files, validate structure, or manage state machines
"I'll skip the gate check โ the handoff looks fine"
Gates catch structural issues your pattern-matching missed
"One long session is faster than spawning agents"
Fresh 200k contexts produce dramatically better output per agent
"I'll just create the message file myself"
You WILL get the numbering wrong eventually. The CLI won't.
Iron Questions
Before starting a council, answer these:
Is this actually complex enough for a council? (If a single session can handle it in <30% context, skip the council.)
Which preset fits? (Don't default to Full โ most tasks need Rapid or Debug.)
Does a Memory Module exist? (If not, the first agent must be a mapper.)
What is the concrete deliverable? (Councils without clear objectives meander.)
# Initialize council with a preset
council init --preset full --objective "Implement user authentication system"# Available presets: full, rapid, debug, architecture, refactoring, audit
This creates .planning/council/ with STATE.md, BOARD.md, and the agent sequence.
If no Memory Module exists (.planning/memory/ is missing or stale >48h):
# Spawn mapper agent to create Memory Module
Task("Map the codebase and create Memory Module files in .planning/memory/.
Scan: project structure, database schemas, API routes, service dependencies,
frontend components, tech stack. Write one file per category.")
Wait for mapper to complete before proceeding. Every subsequent agent needs this context.
Phase 2: Agent Execution Loop
For each agent in the preset sequence:
Step 1 โ Create routing message:
council message --to researcher --task "Analyze authentication patterns in codebase" \
--context "memory/api-routes.md,memory/service-graph.md"
The CLI creates messages/msg-NNN.md with proper numbering and format.
Step 2 โ Spawn the agent:
Task("You are the RESEARCHER agent for this council.
READ these files:
- .planning/council/messages/msg-001.md (your assignment)
- .planning/memory/api-routes.md (context)
- .planning/memory/service-graph.md (context)
EXECUTE your research task as described in the message.
WRITE your findings to: .planning/council/handoffs/handoff-001-researcher.md
Your handoff MUST include these sections:
## Summary
## Findings
## Risks & Concerns
## Recommendations for Next Agent
## Files Examined")
Each agent gets a fresh 200k context. Pass file paths, not file contents.
Step 3 โ Validate the handoff:
council handoff --validate handoff-001-researcher.md
CLI evaluates gate criteria for the current phase. Returns PASS or FAIL with reasons.
If gate fails:
# Option A: Respawn agent with feedback
council message --to researcher --task "Gate failed: missing risk analysis. Revise handoff."# Then respawn agent...# Option B: Force advance (requires user permission)
council advance --force --reason "User approved: risk analysis deferred to architect"
Step 5 โ Advance state:
council advance
Moves state machine to next agent. Only works after gate passes (or --force).
Step 6 โ Regenerate board:
council board
Rebuilds BOARD.md from current state. Shows completed phases, current agent, remaining work.
Repeat steps 1-6 for each agent in sequence.
Phase 3: Review & Synthesis
The final agent (always a reviewer) examines all previous handoffs:
Task("You are the REVIEWER agent for this council.
READ these files:
- .planning/council/messages/msg-NNN.md (your assignment)
- ALL files in .planning/council/handoffs/ (previous agent work)
- Relevant .planning/memory/ files
REVIEW all handoffs for:
- Completeness against the original objective
- Consistency between agents' recommendations
- Gaps or contradictions
- Quality of implementation (if executor was involved)
WRITE your review to: .planning/council/handoffs/handoff-NNN-reviewer.md
Include: ## Verdict (APPROVE / REVISE / REJECT), ## Issues Found, ## Final Recommendations")
After reviewer completes:
council handoff --validate handoff-NNN-reviewer.md
council gate-check --phase review
council advance # Moves to COMPLETE state
council board # Final board showing all phases done
Update persistent memory with council outcomes:
Update .planning/MEMORY.md with decisions made
Update .planning/memory/ if codebase structure changed
Tasks are atomic (<30min each), dependencies explicit, acceptance criteria set
Execution -> Review
All tasks addressed, tests pass, no unresolved TODOs
Review -> Complete
Critical issues resolved, Memory Module updated if needed
Orchestrator Guidelines
The orchestrator (you, the main session) must stay lean:
DO read handoffs to understand progress
DO use CLI commands for all structural operations
DO pass file paths to agents, not file contents
DO keep your own context under 15% โ let agents do the heavy lifting
DON'T do agent work yourself โ spawn a Task()
DON'T hold agent outputs in your context โ they're in files
DON'T manually create or edit any file in .planning/council/
Spawning Agents Effectively
Good agent prompt structure:
Task("You are the [ROLE] agent for this council.
READ: [list of file paths]
EXECUTE: [clear description of what to do]
WRITE: [exact output file path]
FORMAT: [required sections in output]")
Bad agent prompt:
Task("Look at the codebase and figure out what needs to happen for auth.")
Be specific. Agents have fresh contexts โ they know nothing except what you tell them.
Red Flags
Stop and reassess if you observe:
Red Flag
What's Wrong
Single session handling multiple agent roles
You're role-switching, not spawning. Quality will degrade.
Manual file creation in .planning/council/
CLI exists for a reason. State will drift.
No handoff validation between agents
Garbage in, garbage out. Validate before advancing.
Orchestrator context >50%
You're doing agent work. Spawn instead.
Agents referencing work without file paths
Hallucination risk. Always pass explicit paths.
Skipping gate checks
You're trusting vibes over structure.
BOARD.md manually edited
Board is generated output. Use council board.
Error Recovery
Issue
Recovery
Agent produces empty/garbage handoff
Respawn with clearer instructions and more context files
Gate check fails
Read failure reason, fix handoff or --force with user permission
STATE.md corrupted
council init --recover rebuilds from handoffs/ and messages/
Memory Module stale
Spawn mapper agent to refresh before continuing
Agent stuck on wrong task
New routing message with correction, respawn agent
Council objective changed mid-flight
council update --objective "new objective", reassess remaining agents
CLI dependency:planning-tools must be available for all council commands
Quick Reference
# Full lifecycle
council init --preset rapid --objective "Add rate limiting to API"
council message --to researcher --task "..." --context "memory/api-routes.md"# [spawn researcher via Task()]
council handoff --validate handoff-001-researcher.md
council gate-check --phase research
council advance
council board
council message --to executor --task "..." --context "memory/api-routes.md,handoffs/handoff-001-researcher.md"# [spawn executor via Task()]
council handoff --validate handoff-002-executor.md
council gate-check --phase execution
council advance
council board
council message --to reviewer --task "..." --context "handoffs/"# [spawn reviewer via Task()]
council handoff --validate handoff-003-reviewer.md
council gate-check --phase review
council advance
council board
# Done โ update MEMORY.md with outcomes
