| name | or-quick-lookup |
| description | Use when the user mentions an OpenRouter model ID mid-conversation and needs a quick capability/pricing snapshot — not a full evaluation. |
Quick Model Lookup
Retrieve a single model's essential metadata from OpenRouter: pricing, context length, supported modalities, and provider. This is the lightweight skill for rapid checks without deep evaluation.
When to use
- User asks "what is [model-id]?" or "what model is that?"
- User asks about pricing for a specific model ("how much does X cost?")
- User asks a single-attribute question about a model ("is X vision-capable?", "what's the context window?")
- Mid-conversation mention of a model ID that needs quick clarification
Inputs to gather
- Model ID (slash-form, e.g.
anthropic/claude-3.5-sonnet or openai/gpt-4o)
Procedure
- Call
mcp__model-scout__get_model with the model ID.
- Extract and summarize: input pricing per 1M tokens, output pricing per 1M tokens, context length, supported modalities (text, vision, audio, etc.), and provider name.
- Return a 1–2 sentence summary with the key facts.
Output / side effects
User receives pricing (input/output per 1M tokens), context window, modalities, and provider. No persistent artifacts.
If the user wants capability comparisons, benchmarks, or cost projections, suggest or-evaluate-model, or-compare-models, or or-cost-projection respectively.
