Route any prompt to the highest-performing LLM based on thousands of peer-reviewed council deliberations.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install llmcouncil-router
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install llmcouncil-router using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is LLM Council Router?
LLM Council Router is a sophisticated routing engine designed to eliminate the trial-and-error approach to model selection. By leveraging thousands of peer-reviewed council deliberations, this tool identifies the top-performing large language model for any specific query. It provides developers with a data-backed recommendation engine, ensuring that tasks are always handled by the most capable model available.
As part of the ecosystem of Openclaw Skills, this tool acts as an intelligent middleware layer. It doesn't just guess which model is better; it analyzes real evaluation data to return a specific model ID, relevance score, and confidence level, allowing for high-precision automation in complex AI workflows.
LLM Council Router Use Cases
- Selecting the optimal model for specialized tasks such as quantum physics explanations or complex coding challenges.
- Implementing dynamic dispatching in multi-agent systems to ensure the highest quality output for every sub-task.
- Reducing model hallucination by routing prompts to models with proven high performance in specific domains.
- Benchmarking and selecting models via Openclaw Skills to maintain cost-efficiency without sacrificing quality.
How LLM Council Router Works
- The agent or application sends the user's prompt to the LLM Council routing endpoint.
- The system searches its database for the nearest council evaluations related to the specific query type.
- The router calculates model rankings based on nearby wins and appearances in peer-reviewed tests.
- A JSON response is returned containing the top recommended model ID and a confidence score.
- The developer or agent uses this recommendation to call the specific model, often chaining with providers like OpenRouter.
LLM Council Router Setup
To utilize this skill, you must first obtain an API key from the LLM Council developer portal. Once obtained, set your environment variable to authenticate your requests.
export LLMCOUNCIL_API_KEY='clwb_YOUR_KEY_HERE'
You can then trigger a route request using a simple curl command:
curl -X POST https://clawbot.llmcouncil.ai/v1/route \
-H "Content-Type: application/json" \
-H "X-API-Key: $LLMCOUNCIL_API_KEY" \
-d '{"query": "Explain quantum entanglement simply"}'
LLM Council Router Data Schema & Taxonomy
The LLM Council Router returns structured data to facilitate easy integration with other Openclaw Skills and model providers.
| Property | Description |
|---|---|
model |
The recommended model string, formatted for direct use with OpenRouter. |
relevance |
A float (0-1) indicating how closely the query matches historical evaluation data. |
confidence |
A float (0-1) representing the margin by which the top model won against competitors. |
model_rankings |
An array of objects detailing the performance of all considered models. |
LLM Council Router Advanced Features
- Native OpenRouter integration: Model IDs match the OpenRouter catalogue exactly, requiring no mapping code.
- Peer-Reviewed Intelligence: Rankings are derived from human and model-in-the-loop deliberations, not just synthetic benchmarks.
- Adjustable Sensitivity: Use the k-parameter to define how many past evaluations should influence the routing decision.
- High-Scale Throughput: Supports up to 10,000 requests per day for Pro users within the Openclaw Skills framework.
SKILL.md
Loading
Related Openclaw Skills
Odoo Financial Intelligence
A professional-grade financial intelligence tool that generates visual reports and ledger-based analytics directly from Odoo ERP data.
Odoo Financial Intelligence & CFO Agent
A professional read-only AI agent for Odoo that generates ledger-based financial reports, sales analytics, and automated visualizations in PDF, Excel, and WhatsApp formats.
NPkill: Node.js and Next.js Build Artifact Cleaner
A specialized CLI utility to identify and remove bloated node_modules and .next folders to reclaim storage space.
Verify Claims Skill
A professional-grade verification tool that cross-references claims against global fact-checking organizations to combat misinformation and hoaxes.
Pear iCloud Integration
A comprehensive suite of 27 MCP tools that provides read/write access to iCloud Calendar, Reminders, and Contacts through Pear's CalDAV and CardDAV bridge.
PDF OCR using Gemini LLM
A sophisticated AI-powered tool that leverages Google Gemini to perform high-accuracy OCR on scanned or image-based PDF documents.
Featured*
