Engages in a conversation using the Sonar API. Accepts an array of messages (each with a role and content) and returns a chat completion response from the Perplexity model.
AI agents call perplexity_ask to retrieve information from MCP Perplexity Server without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool queries an AI model (Perplexity.ai Sonar API) for conversational responses. It retrieves information and generates text based on input but does not modify, delete, or execute external operations. The only risk is if the conversation itself is used to inform subsequent high-risk actions, but the tool itself is purely a read operation (query + response generation).
From the tool's definition Tool 'engages in a conversation' and 'returns a chat completion response' with no mention of data modification, deletion, or external action triggers.
Attacks that exploit this kind of access
Engages in a conversation using the Sonar API. Accepts an array of messages (each with a role and content) and returns a chat completion response from the Perplexity model. It is categorised as a Read tool in the MCP Perplexity Server MCP Server, which means it retrieves data without modifying state.
Register the MCP Perplexity Server MCP server in PolicyLayer and add a rule for perplexity_ask: 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 Perplexity Server. Nothing to install.
perplexity_ask 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 perplexity_ask 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 perplexity_ask. 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.
perplexity_ask is provided by the MCP Perplexity Server MCP server (samwang0723/mcp-perplexity). 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 →