Zero-shot text classification. POST { text, labels[], multiLabel? }. Assigns the text to one of your labels (or several when multiLabel=true) with a confidence score and a one-line rationale. No training data needed — define the labels at call time. Great for routing, tagging, triage, and intent ...
AI agents call ai.classify to retrieve information from Mcp without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
ai.classify is a stateless inference tool that analyzes and categorizes text. It retrieves classification results based on provided labels but does not create, modify, delete, or execute any operations. The POST method is used for input payload transport, not for state mutation. This is a classic Read operation: query/analyze data with no side effects.
From the tool's definition Tool performs 'Zero-shot text classification' by assigning text to predefined labels and returning confidence scores and rationale.
Attacks that exploit this kind of access
Zero-shot text classification. POST { text, labels[], multiLabel? }. Assigns the text to one of your labels (or several when multiLabel=true) with a confidence score and a one-line rationale. No training data needed — define the labels at call time. Great for routing, tagging, triage, and intent detection. It is categorised as a Read tool in the Mcp MCP Server, which means it retrieves data without modifying state.
Register the MCP server in PolicyLayer and add a rule for ai.classify: 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 Mcp. Nothing to install.
ai.classify is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the ai.classify 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 ai.classify. 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.
ai.classify is provided by the MCP server (@2sio/mcp). 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 →