Use when coordinating team members across agent sessions, assigning roles, managing messaging, and maintaining team status. Trigger with team formation, updates, or coordination needs.
npx claudepluginhub emasoft/emasoft-plugins --plugin emasoft-chief-of-staffThis skill uses the workspace's default tool permissions.
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.
Defines structured protocols for agent team communication: direct messages, broadcasts, plan approvals, shutdowns, and anti-patterns. Use for team norms, message selection, or coordination debugging.
Creates and manages dynamic teams of domain-specific agents. Analyzes project to propose 5-20 agents with tracking framework. Modes: create, update, status, cleanup.
Use when dispatching subagents, composing prompts for teammates, structuring handoff reports, or managing context boundaries between agents. Covers both subagent prompts and team-level messaging.
Share bugs, ideas, or general feedback.
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.
Before using this skill, ensure:
| 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 |
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:
Matching agents to roles based on capabilities and task requirements.
Facilitating communication between team members via AI Maestro.
Maintaining visibility into all team members' current status and activities.
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.
Related documentation:
When to use: When broadcasting updates, sending targeted instructions, or facilitating inter-agent communication.
Steps: Identify recipients, compose message with priority, send via AI Maestro API, confirm delivery, log in coordination state.
Related documentation:
When to use: Continuously during coordination, before assigning tasks, when reporting team status.
Steps: Poll AI Maestro for active sessions, query each agent's status, update team roster, identify inactive agents, flag issues.
Related documentation:
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.
Runbook for composing and sending broadcast or targeted messages to team members via the AI Maestro API, confirming delivery, and logging coordination actions.
Runbook for polling AI Maestro for active sessions, querying agent status, updating the team roster, identifying inactive agents, and flagging issues.
Copy this checklist and track your progress:
Use the agent-messaging skill to send a role assignment message:
helper-agent-genericRole Assignment: Code Reviewerhighrole-assignment, explaining that the agent is assigned the Code Reviewer role with responsibilities: review PRs, enforce code standards, provide feedbackVerify: confirm message delivery and await role acceptance acknowledgment.
Use the agent-messaging skill to broadcast a message to all team members:
Sprint Planning Completenormalannouncement, informing that Sprint 5 planning is complete and all tasks are assignedVerify: confirm delivery to all team members.
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.
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
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."
}
}
Input: Query team status via AI Maestro API
curl -s "http://localhost:23000/api/sessions" | jq '.sessions[] | select(.project == "auth-service")'
Output: Current team status result
[
{"name": "eaa-auth-service-architect", "role": "Architect", "status": "active", "last_seen": "2026-02-14T10:30:00Z"},
{"name": "eoa-auth-service-orchestrator", "role": "Orchestrator", "status": "active", "last_seen": "2026-02-14T10:29:45Z"},
{"name": "auth-service-programmer-001", "role": "Programmer", "status": "active", "last_seen": "2026-02-14T10:28:12Z"},
{"name": "auth-service-programmer-002", "role": "Programmer", "status": "idle", "last_seen": "2026-02-14T10:15:00Z"}
]
Symptoms: No acknowledgment received, agent continues previous behavior.
Solution: Retry with higher priority, check if agent is active, escalate to user if unresponsive after 3 attempts.
Symptoms: Delivery confirmation missing, recipients unaware of messages.
Solution: Verify AI Maestro is running, check recipient session names, validate message format.
Symptoms: Inactive agents shown as active, missing new agents, wrong status.
Solution: Force refresh via status polling, clear and rebuild roster, verify session registry.
Version: 1.0 Last Updated: 2025-02-01 Target Audience: Emasoft Chief of Staff Agent Difficulty Level: Intermediate