AI agents call predict.search 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.
The tool searches and returns prediction market data. While the broader server context involves financial settlement (USDC payments), this specific tool only reads/queries existing prediction market information without moving funds, modifying markets, or executing trades. Search operations are inherently Read-category actions.
From the tool's definition Tool description explicitly states it performs 'keyword search' across prediction markets, returning hits with venue tags. This is a pure retrieval operation with no data modification, deletion, or execution of external operations.
Attacks that exploit this kind of access
Cross-venue prediction-market keyword search (Polymarket + Kalshi + Limitless) with a venue tag on each hit. 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 predict.search: 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.
predict.search 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 predict.search 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 predict.search. 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.
predict.search 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 →