Dispatch prompts to local or cheap LLMs with auto-routing and health checks.
npx skills add https://github.com/isndotbiz/ged --skill orchestrator-isndotbiz
Please help me install this Agent Skill. Skill: orchestrator Source: https://github.com/isndotbiz/ged/tree/main/.claude/skills/orchestrator Command: npx skills add https://github.com/isndotbiz/ged --skill orchestrator-isndotbiz
Dispatch prompts to local/cheap LLMs via the orchestrator to avoid expensive Claude subagent calls and reduce operational costs.
Dispatch a prompt to the best available worker using the orchestrator.
High-intent search queries and answers about installing and using this skill.
To reduce Claude usage costs, you can dispatch prompts to local or cheap LLMs using an orchestrator that auto-routes tasks to the best available workers, cutting operational spend.
LLM prompt routing dispatches tasks to cheaper local models or external workers instead of expensive subagents, tracking spend and applying health checks to minimize operational costs.
Yes, you can use auto-dispatch and swarm capabilities to gather multiple perspectives for code review, generation, research, and validation tasks by routing prompts to available workers.
You can track LLM spend and monitor worker availability by querying the health and spend endpoints, which inform budgeting decisions and check the status of routed tasks.
Yes, the prompt dispatch approach works with OpenRouter, CLI agents, GLM, and TabbyAPI, routing prompts to these workers to minimize expensive Claude subagent calls.
Routing prompts to cheap LLMs limits result quality compared to expensive subagents, though auto-dispatch and swarm capabilities help gather multiple perspectives to improve outputs.
citedyConfigure LLM providers, models, and API keys via a local management API.
patricio0312revOptimize LLM applications for cost and latency via caching, model selection, batching, and prompt optimization.
ylxmf2005Verify code review findings and issue verdicts with root-cause analyses and fixes.
Chair4ceExecute parallel and batched tasks using distributed Gemini Flash workers.
