Onboard Swarm

Description

--- description: The parallel per-agent swarm that beats cold-start — each bounded agent reads its canonical source (git/claude-code → Analyst, granola → SDR, cross-stream → Auditor) in parallel and posts a candidate (the one fact, commitment, or risk worth acting on) before any deliberation. Read-only, LLM-optional. --- Invoke the `onboard-swarm` skill and follow its SKILL.md flow exactly: run the swarm, render the candidates grouped by the bounded agent that posted each (Analyst / SDR / Audit

Installation

Installs to ~/.claude/skills/liminal-space-io-liminal-agents-onboard-swarm/SKILL.md

Restart Claude Code, or start a new session, for it to be picked up.