{"skill":{"slug":"cerberus-agent-foundry","displayName":"Isolated Multi-Agent Control Plane","summary":"Production-ready multi-agent architecture kit for OpenClaw. Provides isolated per-agent workspaces, control-plane orchestration, structured task lifecycle, c...","description":"---\nname: cerberus-agent-foundry\ndescription: \"Production-ready multi-agent architecture kit for OpenClaw. Provides isolated per-agent workspaces, control-plane orchestration, structured task lifecycle, checksum-verified mailbox messaging, and auditable event logs. Use when implementing Team Lead + specialist agents with secure boundaries, async collaboration, and operational governance.\"\n---\n\n# Cerberus Multi-Agent Control Plane\n\nA professional blueprint for building, operating, and governing local multi-agent teams in OpenClaw. It ships a hardened protocol stack for role isolation, task orchestration, asynchronous handoffs, and auditability.\n\n## Quick Start\n\n1. Install blueprint into target path:\n   - `bash scripts/install_blueprint.sh /path/to/your/system`\n2. Verify task board:\n   - `python3 /path/to/your/system/scripts/taskctl.py list`\n3. Verify mailbox:\n   - send: `python3 /path/to/your/system/scripts/mailboxctl.py send --task-id TSK-000 --sender team-lead --receiver coder --correlation-id CORR-001 --body \"ACK protocol\"`\n   - ack: `python3 /path/to/your/system/scripts/mailboxctl.py status --message-id MSG-0001 --to ACK --actor coder`\n\n## What This Blueprint Provides\n\n- Strict workspace isolation per agent (`agents/*/workspace`)\n- Unified control plane (`control-plane/tasks`, `control-plane/mailbox`, `control-plane/logs`)\n- Task state machine tooling (`taskctl.py`)\n- Mailbox protocol tooling with checksum and GC (`mailboxctl.py`)\n- Shared memory layer under control plane (`control-plane/shared-memory`)\n\n## Core Rules (must enforce)\n\n1. Never use shared agent workspaces.\n2. Use only `control-plane/mailbox` for agent-to-agent messaging.\n3. Team Lead is sole mailbox garbage-collection authority.\n4. Deploy actions require explicit human approval artifacts.\n\n## References\n\n- Read `references/operations-checklist.md` for rollout and audit checks.\n- Read `assets/blueprint/docs/protocol-v1.md` for protocol details.\n","topics":["Multi Agent","Orchestration","Messaging"],"tags":{"latest":"0.1.2"},"stats":{"comments":0,"downloads":653,"installsAllTime":24,"installsCurrent":0,"stars":0,"versions":3},"createdAt":1772304741175,"updatedAt":1778491670926},"latestVersion":{"version":"0.1.2","createdAt":1772305104607,"changelog":"Rename project title to a function-first professional name; keep architecture and protocol stack unchanged.","license":null},"metadata":null,"owner":{"handle":"delibrately-cmyk","userId":"s174555bk480w0q4adm04g0ta9884vyq","displayName":"delibrately-cmyk","image":"https://avatars.githubusercontent.com/u/253489992?v=4"},"moderation":null}