Compound research: search → fetch top pages → summarize → synthesize.
AI agents call research to retrieve information from mcp Research without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool performs information retrieval and synthesis across multiple sources. While it orchestrates multiple operations (searching, fetching, summarizing), all operations are read-only with no data modification, deletion, or execution of arbitrary code. The server explicitly describes these as read-only tools with SSRF protection, indicating safe content fetching.
From the tool's definition The tool description indicates "search → fetch top pages → summarize → synthesize" with server-level description stating "read-only tools".
Attacks that exploit this kind of access
Compound research: search → fetch top pages → summarize → synthesize. It is categorised as a Read tool in the mcp Research MCP Server, which means it retrieves data without modifying state.
Register the mcp Research MCP server in PolicyLayer and add a rule for 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 mcp Research. Nothing to install.
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 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 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.
research is provided by the mcp Research MCP server (mabaam/maibaamcrawler). 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 →