From ai-maestro-programmer-agent
Communication with AMOA Orchestrator via AI Maestro. Use when sending clarifications, status updates, blockers, or completions. Trigger with /ampa-orchestrator-comm. Loaded by ai-maestro-programmer-agent-main-agent.
How this skill is triggered — by the user, by Claude, or both
Slash command
/ai-maestro-programmer-agent:ampa-orchestrator-communicationai-maestro-programmer-agent-main-agentThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Defines all communication protocols between AMPA and AMOA. Uses asynchronous
Defines all communication protocols between AMPA and AMOA. Uses asynchronous
inter-agent messaging via the globally installed agent-messaging skill. Read
that skill once during initialization to learn messaging commands and syntax.
Copy this checklist and track your progress:
agent-messaging skill and follow its
initialization to register your messaging identity.type, priority,
subject, and structured content as specified in the reference.Sent/received messages to/from AMOA via agent-messaging skill.
If message delivery fails after 3 retries, write the message content to
docs_dev/unsent-<timestamp>.md and report the delivery failure to your local
log. Resume when connectivity is restored.
{"type": "status-update", "message": "Task #42: 3/5 criteria met. Details: docs_dev/status-42.md"}| Document | Description |
|---|---|
| op-comprehension-handshake.md | The task-comprehension handshake answer (all 5 points, before coding) — replaces the bare ACK (R6 v3 / #17 M7a) |
| op-request-clarification.md | When to Use, Prerequisites, Procedure, Examples, Error Handling |
| op-report-status.md | When to Use, Prerequisites, Procedure, Examples, Error Handling |
| op-report-blocker.md | When to Use, Prerequisites, Procedure, Examples, Error Handling |
| op-propose-improvement.md | When to Use, Prerequisites, Procedure, Examples, Error Handling |
| op-notify-completion.md | When to Use, Prerequisites, Procedure, Examples, Error Handling |
| op-pre-pr-gate.md | The "PR now?" pre-PR gate — ask AMOA for a green-light before opening a PR (R6 v3 / #17 M7c) |
| op-receive-feedback.md | When to Use, Prerequisites, Procedure, Examples, Error Handling |
agent-messaging skill (global) — Messaging commands.
This skill operates under the AI Maestro approval tiers — the escalation
ladder Tier 0 (self-authority for in-scope work + DERIVED NPT/EHT tasks) →
CHIEF-OF-STAFF → MANAGER → USER. See the ampa-prrd-trdd-kanban skill and
~/.claude/rules/trdd-approval-tiers.md for which transitions need approval.
A MEMBER never self-approves its own releases — entering the release
pipeline (publish/deploy) is USER/MANAGER-authorized, and the INTEGRATOR
(not the MEMBER) owns the → complete flip.
npx claudepluginhub emasoft/ai-maestro-programmer-agentWhole-repo audit for over-engineering: finds dead code, unnecessary abstractions, stdlib-replaceable dependencies. Outputs ranked findings and net line/dep savings.