Send a message to any of 40 AI models. Returns the model's response. Supports conversation history via the messages array.
| Name | Required | Description | Default |
|---|---|---|---|
| model | No | Model to use, e.g. 'claude-sonnet-4.6', 'gpt-4.1', 'llama-3.1-8b'. Defaults to llama-3.2-1b | |
| messages | Yes | Array of chat messages, each with 'role' (user/assistant/system) and 'content' (text) | |
| max_tokens | No | Maximum tokens to generate in the response. Default 1024 |
Changes observed during successful MCP inspections.
Input schema / properties / model / descriptionPrevious value: -"Model to use, e.g. 'claude-sonnet-4.6', 'gpt-4.1', 'llama-3.1-8b'. Defaults to llama-3.1-8b"New value: +"Model to use, e.g. 'claude-sonnet-4.6', 'gpt-4.1', 'llama-3.1-8b'. Defaults to llama-3.2-1b"Input schema / properties / model / enumAdded value: +[
+ "llama-3.2-1b",
+ "llama-3.2-3b",
+ "llama-3.1-8b",
+ "llama-3.1-8b@workers-ai",
+ "gemini-2.5-flash-lite",
+ "gemini-2.0-flash",
+ "gpt-oss-20b",
+ "mistral-7b",
+ "glm-4.7-flash",
+ "qwen3-30b",
+ "gpt-4.1-nano",
+ "gpt-4.1-mini",
+ "gpt-4o-mini",
+ "gemini-2.5-flash",
+ "llama-3.3-70b",
+ "llama-3.3-70b@workers-ai",
+ "deepseek-r1-32b",
+ "claude-haiku-4.5",
+ "gpt-oss-120b",
+ "grok-4.1-fast",
+ "grok-3-mini",
+ "glm-5.2",
+ "gemma-4-26b",
+ "mistral-small-3.1-24b",
+ "qwen2.5-coder-32b",
+ "gpt-4.1",
+ "gpt-4o",
+ "o4-mini",
+ "o3-mini",
+ "o3",
+ "claude-sonnet-4.6",
+ "claude-opus-4.6",
+ "claude-opus-4.7",
+ "claude-opus-4.8",
+ "grok-3",
+ "kimi-k2.7-code",
+ "qwen3.8-27b",
+ "gpt-6-luna",
+ "gpt-6.1-sol",
+ "claude-sonnet-5.5",
+ "claude-opus-5.5",
+ "gemini-3.8-flash"
+]Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With readOnlyHint=true already declaring a non-mutating operation, the description adds that a response is returned and that multi-turn context is carried in messages. It omits anything about rate limits, cost, or model-selection tradeoffs, which are the real behavioral unknowns for a 40-model chat endpoint.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Three short sentences, front-loaded with capability and model breadth, followed by return behavior and history support. Slightly wasteful in that the return statement is nearly self-evident, but nothing is padded.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
No output schema exists, and the description does state that the model's response is returned, so the return contract is covered. For a simple three-parameter chat call with full schema coverage, the only missing piece is operational context such as cost or latency differences between models.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100%, so the model enum, messages structure, and max_tokens default are already documented. The description only restates that conversation history is passed through the messages array, adding no syntax or selection guidance beyond the schema.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
States a specific verb ('Send a message') and resource ('any of 40 AI models') plus the outcome ('Returns the model's response'). Sibling tools are all different modalities (embeddings, images, speech, translate), so the purpose is unambiguous even though no sibling is named explicitly.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Usage is implied by the naming of conversation history via the messages array, but there is no explicit when-to-use guidance, no mention of cost/latency tradeoffs across 40 models, and no exclusions. An agent must infer that this is the general text-chat entry point.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Add one secure layer between your agents and this server.