Use when coordinating team members across agent sessions, assigning roles, managing messaging, and maintaining team status. Trigger with team formation, updates, or coordination needs.
Installer avec Codex ou Claude Copiez ce prompt, collez-le dans Codex, Claude ou un autre assistant, puis laissez-le vérifier la page du skill et l'installer pour vous.
Une commande directe contourne le prompt de vérification. Examinez la source avant de l'exécuter.
Use when coordinating team members across agent sessions, assigning roles, managing messaging, and maintaining team status. Trigger with team formation, updates, or coordination needs.
user-invocable
false
license
Apache-2.0
compatibility
Requires AI Maestro messaging API access, session name resolution, and inter-agent communication capabilities. Requires AI Maestro installed.
metadata
{"author":"Emasoft","version":"1.0.0"}
context
fork
agent
ecos-main
workflow-instruction
Step 5
procedure
proc-notify-team-ready
Emasoft Chief of Staff Team Coordination Skill
Overview
Team coordination is a core responsibility of the Chief of Staff agent. This skill teaches you how to manage distributed agent teams, assign roles effectively, coordinate messaging between team members, and maintain awareness of all active teammates and their current status.
Prerequisites
Before using this skill, ensure:
Team registry is accessible
AI Maestro messaging is available
Agent roles are defined
Instructions
Identify coordination need
Query team registry for current state
Execute coordination action
Update team registry and notify agents
Output
Coordination Type
Output
Team formation
New team created, agents assigned
Role assignment
Agent roles updated, registry modified
Status sync
All agents aligned on current state
What Is Team Coordination?
Team coordination is the process of organizing multiple AI agents to work together on shared goals. The Chief of Staff acts as the central coordination point, ensuring agents are assigned appropriate roles, can communicate effectively, and remain aware of each other's progress and status.
Key characteristics:
Distributed: Team members run in separate Claude Code sessions
Asynchronous: Communication happens via message queues, not direct calls
Role-based: Each agent has a defined role with specific responsibilities
Status-aware: Chief of Staff tracks all team member states
Team Coordination Components
1. Role Assignment
Matching agents to roles based on capabilities and task requirements.
2. Team Messaging
Facilitating communication between team members via AI Maestro.
3. Teammate Awareness
Maintaining visibility into all team members' current status and activities.
Core Procedures
PROCEDURE 1: Assign Agent Roles
When to use: When onboarding a new agent, reassigning responsibilities, or restructuring the team.
Steps: Identify required role, match agent capabilities, send role assignment message, confirm acceptance, update team roster.
These operational runbooks provide step-by-step instructions for executing each team coordination procedure. Use them as quick-reference guides when performing the corresponding operation.
Runbook for identifying required roles, matching agent capabilities, sending role assignment messages, confirming acceptance, and updating the team roster.
Steps - Identify required role, match agent capabilities, send role assignment message, confirm acceptance, update team roster
Examples - Assigning code reviewer role, reassigning role during restructure
Error Handling - What to do when role assignment fails
Checklist - Verification steps after role assignment
Runbook for composing and sending broadcast or targeted messages to team members via the AI Maestro API, confirming delivery, and logging coordination actions.
Steps - Identify recipients, compose message with priority, send message, confirm delivery, log in coordination state
Examples - Broadcasting sprint start, targeted instruction to developer, multi-recipient status request
Error Handling - What to do when message delivery fails
Runbook for polling AI Maestro for active sessions, querying agent status, updating the team roster, identifying inactive agents, and flagging issues.
Steps - Poll AI Maestro for active sessions, query each agent's status, update team roster, identify inactive agents, flag issues
Examples - Initial team state discovery, pre-task assignment check, team status report generation
Error Handling - What to do when status polling fails
Checklist - Verification steps after status update
Task Checklist
Copy this checklist and track your progress:
Understand team coordination purpose and components
Learn PROCEDURE 1: Assign agent roles
Learn PROCEDURE 2: Send team messages
Learn PROCEDURE 3: Maintain teammate awareness
Practice role assignment workflow
Practice team messaging via AI Maestro
Practice status monitoring and reporting
Examples
Example 1: Assigning a Role to a New Agent
Use the agent-messaging skill to send a role assignment message:
Recipient: helper-agent-generic
Subject: Role Assignment: Code Reviewer
Priority: high
Content: type role-assignment, explaining that the agent is assigned the Code Reviewer role with responsibilities: review PRs, enforce code standards, provide feedback
Verify: confirm message delivery and await role acceptance acknowledgment.
Example 2: Broadcasting a Team Update
Use the agent-messaging skill to broadcast a message to all team members:
Subject: Sprint Planning Complete
Priority: normal
Content: type announcement, informing that Sprint 5 planning is complete and all tasks are assigned
Verify: confirm delivery to all team members.
Example 3: Checking Team Status
Use the ai-maestro-agents-management skill to list all active sessions and their status, including each agent's name, status, and last seen timestamp.
Verify: all expected team members appear in the list with correct status.
Example 4: Full Coordination Workflow with Input/Output
This example shows a complete team coordination cycle with concrete inputs and outputs.
Input: User requests a new team for project "auth-service"
User message: "Set up a team for the auth-service project.
I need an architect, an orchestrator, and two programmers."
Output: ECOS creates the team and confirms formation
[TEAM-FORMED] auth-service
Agents spawned:
- eaa-auth-service-architect (Architect) - ACTIVE
- eoa-auth-service-orchestrator (Orchestrator) - ACTIVE
- auth-service-programmer-001 (Programmer) - ACTIVE
- auth-service-programmer-002 (Programmer) - ACTIVE
Registry: .emasoft/team-registry.json updated
Messages sent: 4 role assignments delivered
Example 5: Role Assignment with Input/Output
Input: ECOS assigns a code reviewer role to an existing agent
{"to":"eia-feature-reviewer","subject":"Role Assignment: Code Reviewer for auth-service","priority":"high","content":{"type":"role-assignment","message":"You are assigned as Code Reviewer for auth-service. Responsibilities: review all PRs, enforce code standards, verify test coverage above 80%."}}
Output: Agent acknowledges the role assignment
{"from":"eia-feature-reviewer","subject":"Role Accepted: Code Reviewer for auth-service","content":{"type":"role-acceptance","message":"Role accepted. Ready to review PRs for auth-service."}}