Setup & Installation
openclaw skills install @edmonddantesj/aoineco-squad-dispatch
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
npx clawhub install aoineco-squad-dispatch
What This Skill Does
Routes tasks across a pre-configured 7-agent squad based on skills, cost, and availability. Separates parallel from sequential tasks automatically and picks the cheapest capable agent for each job. Built on pure Python with no external dependencies.
Unlike sending all tasks to one model, it selects the cheapest agent that meets the task's skill requirements, reducing API costs without manual routing logic.
When to use it
- Routing security audits to a dedicated security agent instead of a general-purpose one
- Running community posts on cheap Gemini Flash while reserving Opus for governance decisions
- Batching independent tasks to run in parallel across multiple agents
- Tracking per-task cost estimates before committing to a dispatch plan
- Managing a mixed-model squad where different agents have different rate limits
Example Workflow
Here's how your AI assistant might use this skill in practice.
INPUT
User asks: run a security audit and post a community update after it completes
AGENT
- 1Create task TASK-0001 for security audit, assign to Blue-Blade (claude-sonnet, security skill)
- 2Create task TASK-0002 for community post, assign to Blue-Sound (gemini-flash, community skill), mark depends_on TASK-0001
- 3Call dispatcher.plan() to resolve dependency order and group parallel vs sequential tasks
- 4Print formatted dispatch plan showing agent assignments, cost tiers, and execution order
- 5Execute plan, Blue-Blade runs first, Blue-Sound runs after TASK-0001 completes
OUTPUT
Security audit completed by Blue-Blade, community post published by Blue-Sound in dependency order