Request a web page, return the paper from that web page
AI agents call fetch_paper to retrieve information from McpDeepResearch without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool fetches and retrieves paper content from a web page, which is a read-only operation. It has no side effects—it does not create, modify, delete, or execute anything. The action is purely informational retrieval of academic papers. Severity is low because the blast radius of misuse (e.g., fetching unintended papers) is limited to information disclosure without destructive or financial impact.
From the tool's definition Tool name 'fetch_paper' and description 'Request a web page, return the paper from that web page' indicate retrieval of content with no modification or deletion.
Attacks that exploit this kind of access
Request a web page, return the paper from that web page. It is categorised as a Read tool in the McpDeepResearch MCP Server, which means it retrieves data without modifying state.
Register the McpDeepResearch MCP server in PolicyLayer and add a rule for fetch_paper: 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 McpDeepResearch. Nothing to install.
fetch_paper 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 fetch_paper 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 fetch_paper. 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.
fetch_paper is provided by the McpDeepResearch MCP server (pb-207/mcp-deep-research). 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 →