AI Maestro message templates for emasoft plugins. Use when sending task assignments, status reports, or escalations between agents. Trigger with messaging requests.
This skill provides shared AI Maestro message templates and communication protocols used across all emasoft plugins for agent coordination, task assignment, status reporting, and escalation.
agent-messaging skillCopy this checklist and track your progress:
Message Sending Workflow:
agent-messaging skillAll AI Maestro messages use this format:
Note: Use the
agent-messagingskill to send messages. The JSON structure below shows the message content.
{
"from": "<sender-agent-name>",
"to": "<recipient-agent-name>",
"subject": "<short-subject-line>",
"priority": "high|normal|low",
"content": {
"type": "request|response|notification|acknowledgment",
"message": "<human-readable-message>",
"data": {
"task_id": "<optional-task-identifier>",
"pr_number": "<optional-pr-number>",
"issue_number": "<optional-issue-number>",
"status": "<optional-status>"
}
}
}
Send messages using the agent-messaging skill. Provide the JSON payload with recipient, subject, priority, and content fields as described above.
Check your inbox using the agent-messaging skill. Retrieve all unread messages for your session and process the content of each message.
Verify: confirm all messages are delivered or received as expected.
For complete JSON templates with all fields, see references/message-templates.md:
For AI Maestro curl command templates with all message types, see references/ai-maestro-message-templates.md:
For response templates FROM other agents TO EOA, see references/ecos-response-templates.md:
For session lifecycle messages (wake/hibernate/terminate), see references/session-lifecycle-templates.md:
For task lifecycle commands (cancel/pause/resume/broadcast/stop), see references/task-lifecycle-templates.md:
For agent resource and skill requests, see references/resource-request-templates.md:
USER
↓
EAMA (Assistant Manager) - User's interface, approval authority
↓
ECOS (Chief of Staff) - Agent lifecycle, team management
↓ ↓ ↓
EAA (Architect) EOA (Orchestrator) EIA (Integrator)
| From | To | Purpose |
|---|---|---|
| EAMA | ECOS | Project creation, approval decisions, status requests |
| ECOS | EAMA | Approval requests, status reports, escalations |
| ECOS | EOA | Agent availability notifications, team assignments |
| ECOS | EAA | Design requests (via EOA typically) |
| EOA | EAA | Design requests, requirements handoff |
| EOA | EIA | Integration/review requests |
| EOA | Remote Agents | Task assignments, status requests |
| EAA | EOA | Design handoffs |
| EIA | EOA | Integration results, quality reports |
| Any Agent | ECOS | Escalations, resource requests |
All plugins use assign: prefix for agent assignment labels:
# Assign task to agent
gh issue edit <number> --add-label "assign:<agent-name>"
# Query agent's tasks
gh issue list --label "assign:<agent-name>"
| Label | Meaning |
|---|---|
status:todo | Not started |
status:in-progress | Being worked on |
status:blocked | Blocked, needs attention |
status:review | Ready for review |
status:done | Complete |
When multiple agents need to modify the same resources (labels, issues), follow conflict resolution protocol.
See references/conflict-resolution.md for:
Escalation is based on order, priority, and state transitions - not fixed time intervals (agents may hibernate for days).
See references/escalation-protocol.md for:
All emasoft plugins follow consistent record-keeping standards for delegation logs, status reports, and handoff documents.
See references/record-keeping.md for:
docs_dev/ directory structure for each plugin (orchestration, integration, design, chief-of-staff, projects, reports)See references/examples.md for:
| Output Type | Format | Example |
|---|---|---|
| AI Maestro message | JSON | Task assignment, status request, approval |
| Message confirmation | API response | {"status": "sent", "message_id": "xyz"} |
| Message history | JSON array | All messages for an agent |
| Delegation log entry | Markdown | Timestamped record of message sent |
| Error | Cause | Solution |
|---|---|---|
| Message not delivered | AI Maestro offline or agent not found | Check AI Maestro health, verify agent session name |
| No response from agent | Agent hibernated, offline, or unresponsive | Follow escalation order from section 3.6 |
| Invalid JSON | Malformed message content | Validate JSON syntax before sending |
| Wrong recipient | Incorrect agent name or session ID | Verify agent name from roster or AI Maestro |
| Label conflict | Multiple agents modifying same issue | Follow conflict resolution protocol from section 3.5 |
| Scenario | Template | Priority |
|---|---|---|
| Assign task | 2.1 | high |
| Task complete | 2.2 | normal |
| Status request | 2.3 | normal |
| Status response | 2.4 | normal |
| Approval request | 2.5 | high |
| Approval response | 2.6 | high |
| Escalation | 2.7 | high |
| Acknowledgment | 2.8 | low |
| Design handoff | 2.9 | high |
| Integration request | 2.10 | high |
| Integration result | 2.11 | high |