一键导入
msg
Inter-agent messaging — send tasks, questions, and decisions to other Mori agents over NATS
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
菜单
Inter-agent messaging — send tasks, questions, and decisions to other Mori agents over NATS
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
| name | msg |
| description | Inter-agent messaging — send tasks, questions, and decisions to other Mori agents over NATS |
Parse the user's input:
send <to> <type> <body> — call mori-msg_send(to=<to>, type=<type>, body=<body>)recv or inbox — call mori-msg_recv(unacked=True), format as inboxthread <id> — call mori-msg_thread(id=<id>)send --broadcast <body> — call mori-msg_send(to="broadcast", type="broadcast", body=<body>)ack <id> — call mori-msg_send(to="<from_host>", type="ack", reply_to=<id>, body="acknowledged")done <id> — call mori-msg_send(to="<from_host>", type="done", reply_to=<id>, body="completed")Valid types: task, decision, question, reply, ack, done, broadcast.
For tasks and questions in inbox output, include a ready-made reply command so the agent can respond in one step.
Example inbox format:
── Pending Messages ─────────────────────────────────────────────
[task] from laptop 2026-05-31 14:22 id=a3f9c2b1
Extract rate limiting into its own middleware module in src/middleware/
→ ack: mori-msg_send(to="laptop", type="ack", reply_to="a3f9c2b1-...", body="acknowledged")
[question] from gce 2026-05-31 15:01 id=b4e8f3c2
Did you resolve the JWT expiry edge case in session.rs?
→ reply: mori-msg_send(to="gce", type="reply", reply_to="b4e8f3c2-...", body="...")
Strategic guidance from the advisor model on a question, with optional focus area and depth. Use before consequential decisions or to review a plan.
Session bootstrap — loads shared memories and team standards from the Mori server. Use at session start, or after a context compaction with --post-compact.
Bundle the canonical memory set into one structured Markdown document for external-LLM review, audit, or dashboard download.
Ranked full-text search and browse over the shared memory store. Use to recall past decisions, patterns, or project context.
Runs the dream pipeline that distils session events into durable memories. Use to flush undreamed events or check dream status.
Bootstraps the memory store from existing material (repos, PDFs, transcripts). Use to seed memory from a project or document.