Conduct deep, multi-source research on a topic (Sonar Deep Research model).
AI agents call perplexity_research to retrieve information from Perplexity API Platform MCP Server without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves and aggregates information from multiple web sources to present research findings. It does not create, modify, delete, execute code/commands, or move money. It is a read-only information-gathering operation analogous to a search or fetch operation. The low severity reflects minimal blast radius—misuse would only return unwanted information, not cause side effects or damage.
From the tool's definition The tool description states it 'conduct[s] deep, multi-source research on a topic' using the Sonar Deep Research model.
Documented attack patterns abuse exactly the kind of access perplexity_research gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Perplexity API Platform MCP Server, and nothing reaches the server without passing your rules. This is the rule we recommend for perplexity_research:
{
"version": "1",
"default": "deny",
"tools": {
"perplexity_research": {}
}
}perplexity_research is read-only, so it stays allowed — but everything else on the server is denied unless you say otherwise.
Free to start. No card required.
Conduct deep, multi-source research on a topic (Sonar Deep Research model). It is categorised as a Read tool in the Perplexity API Platform MCP Server MCP Server, which means it retrieves data without modifying state.
Register the Perplexity API Platform MCP Server MCP server in PolicyLayer and add a rule for perplexity_research: 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 Perplexity API Platform MCP Server. Nothing to install.
perplexity_research 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_research 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_research. 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_research is provided by the Perplexity API Platform MCP Server MCP server (perplexityai/modelcontextprotocol). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Deterministic rules across all 4 Perplexity API Platform MCP Server tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.
Free to start. No card required.
4 Perplexity API Platform MCP Server tools catalogued and risk-classified — across an index of 42,500+ MCP servers.