AI team intelligence and fleet management for OpenClaw. Spawn, monitor, and orchestrate a fleet of AI lobster assistants in Docker — with proactive bottleneck detection, peer-to-peer knowledge sharing, health checks, rolling upgrades, and owner notifications.
Distributed goal orchestration system. Decomposes big goals into sub-goals, distributes to lobsters, monitors progress via event-driven + patrol fallback, and verifies completion. Integrates with task boards (Notion/Linear/GitHub) and messaging (Slack/Discord).
Structured workflow for collaborating with ACP coding agents (Claude Code, Codex, Cursor, etc). Confidence-based routing, structured prompts, output verification, and session management. Ensures code changes are correct, tested, and reviewed before merging.
Proactive task discovery engine. Periodically scans task boards (OKR, GitHub Issues, Notion, Linear) for items the lobster can pick up, generates structured approval requests with confidence scoring. Turns lobsters from reactive responders into self-driving agents.
Automatic task decomposition and execution planning. Receives a high-level task, breaks it into subtasks, classifies each (code/research/design/ops), estimates effort, identifies dependencies, and generates a structured plan for owner approval. Works standalone or as part of goal-ops workflows.
Generate Excalidraw diagrams from natural language descriptions. Use when asked to "create a diagram", "make a flowchart", "visualize a process", "draw a system architecture", "create a mind map", or "generate an Excalidraw file". Supports flowcharts, relationship diagrams, mind maps, and system architecture diagrams. Outputs .excalidraw JSON files that can be opened directly in Excalidraw.
Automatically learn from corrections, errors, and reflections. Triggers when you're corrected, make mistakes, discover better approaches, or complete important tasks. Uses semantic detection — not keyword matching.
Universal approval flow framework — receives structured proposals, sends to Slack channel/DM/thread, parses human approval commands (execute/partial execute/skip/reject), callbacks to execution chain. All scenario SKILLs requiring human approval should call this atomic capability, don't implement approval logic yourself.