name: julia-openclaw-token-optimizer
description: Julia's OpenClaw Token Optimizer: Scans LLM prices, benchmarks models, patches config for cheapest optimal. Use for cost optimization, model rotation.
Usage
- session_status → current model/cost.
- web_search 'current LLM pricing [provider] input/output per million tokens'.
- gateway config.schema.lookup 'agents.defaults.modelSelection'.
- Benchmark: spawn subagent with test prompts on cheap models.
- Recommend: Cost/speed/quality matrix.
- gateway config.patch {modelSelection: {primary: 'best-cheap-model'}}.
References: providers.md for price APIs.
Optimization Rules
- Prefer fast/cheap for simple tasks.
- Quality for creative/complex.
- Rotate on rate limits.
Example: 'Optimize my config for cheap reasoning' → Search prices, patch.