DashClaw platform expert for integration, troubleshooting, and governance. Snapshot-based โ always prefer live queries via `python -m livingcode query`.
Governance behavior for AI agents governed by DashClaw. Teaches the governance protocol: when to call guard (risk thresholds), how to interpret decisions (allow/warn/block/require_approval), when to record actions, how to wait for approvals, and session lifecycle management. Loads org-specific policies and capabilities from MCP resources at session start. Use with @dashclaw/mcp-server. Trigger on: governed agent, dashclaw governance, guard policy, approval wait, governed capability, risk threshold, action recording, session lifecycle.