Provides CLI-style access to a self-hosted SwarmClaw agent runtime. The skill maps to the SwarmClaw REST API and exposes commands for listing and managing agents, creating tasks, chat sessions, schedules, memory entries, goals, knowledge sources, and system health checks. It prefers machine-readable JSON output for automation.
Use this skill when you need to dispatch work to agents, inspect fleet status, create or manage schedules, query memory, run diagnostics, or administrate SwarmClaw resources from an agent. It’s the primary interface for operational control of a SwarmClaw instance.
Works with agents that can run shell commands or call external CLIs and parse JSON output (OpenClaw, CLI-capable agents, and automation scripts).
SwarmClaw is a CLI skill for managing AI agent fleets — agents, tasks, chats, schedules, memory, goals, connectors, and system health via a JSON-friendly CLI mapping to a REST API. No bundled scripts to test. The SKILL.md is comprehensive with clear command examples, output modes, and practical rules. Shell injection risk via --data JSON string interpolation is mitigated by recommending --json output.
Well-structured CLI skill with good security practices (env var for secrets, confirmation before destructive ops, no pipe-to-shell). Long SKILL.md could benefit from splitting command reference into references/ directory. No scripts to audit.