AI agents invoke togetherai_chat_completion to trigger actions in UnClick. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call — builds kicked off, notifications sent, workflows started.
| Parameter | Type | Required | Description |
|---|---|---|---|
stop | array | — | Stop sequences |
model | string | Yes | Model ID (e.g. meta-llama/Llama-3-8b-chat-hf). Use togetherai_list_models to browse. |
top_k | number | — | Top-k sampling |
top_p | number | — | Top-p nucleus sampling |
api_key | string | Yes | Together AI API key |
messages | array | Yes | Array of {role, content} message objects |
max_tokens | number | — | Maximum tokens to generate |
temperature | number | — | Sampling temperature 0-2 (default 0.7) |
Parameters from the server's own tool schema.
This tool executes code/operations by invoking third-party AI models whose behavior depends entirely on the input arguments (prompts). An AI agent could use it to run arbitrary inference operations, generate malicious content, or chain outputs as inputs to other destructive tools.
From the tool's definition The tool description explicitly states 'Run a chat completion with any Together AI model' - the verb 'Run' combined with 'any' model indicates execution of arbitrary operations.
Risk signalsHandles credentials or secrets (api_key)
Attacks that exploit this kind of access
Run a chat completion with any Together AI model. Supports Llama, Mistral, Qwen, and 100+ open-source models. It is categorised as a Execute tool in the UnClick MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
togetherai_chat_completion accepts 8 parameters: stop, model, top_k, top_p, api_key, messages, max_tokens, temperature. Required: model, api_key, messages. The full parameter table on this page comes from the server's own tool schema.
Register the UnClick MCP server in PolicyLayer and add a rule for togetherai_chat_completion: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches UnClick. Nothing to install.
togetherai_chat_completion is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.
Yes. Add a rate_limit block to the togetherai_chat_completion rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for togetherai_chat_completion. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
togetherai_chat_completion is provided by the UnClick MCP server (@unclick/mcp-server). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Every MCP server has a record like this.
Type a name, get the same breakdown: verified identity, auth posture, risk grade, capabilities, recommended policy.
Teams ship this data inside their own products. See what a licence covers →