CreateMessageRequestParams

A request to include context from one or more MCP servers (including the caller), to be attached to the prompt. The client MAY ignore this request.

The requested maximum number of tokens to sample (to prevent runaway completions). The client MAY choose to sample fewer tokens than the requested maximum.

The messages to use as context for sampling. Typically includes conversation history and the current user message.

Optional metadata for this request. May include a progressToken for out-of-band progress notifications.

Optional metadata to pass through to the LLM provider. The format of this metadata is provider-specific.

The server's preferences for which model to select. The client MAY ignore these preferences and choose any model.

Optional list of sequences that will stop generation if encountered.

An optional system prompt the server wants to use for sampling. The client MAY modify or omit this prompt.

Optional temperature parameter for sampling (typically 0.0-2.0). Higher values make output more random, lower values make it more deterministic.

Optional policy controlling how the model uses the provided tools. The client MUST return an error if this field is present but the client did not advertise ClientCapabilities.Sampling.tools.

Optional list of tools the model may use during generation. The client MUST return an error if this field is present but the client did not advertise ClientCapabilities.Sampling.tools.