Read the full content of a stored paper in markdown format
AI agents call read_paper to retrieve information from ArXiv 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 displays paper content without creating, modifying, deleting, executing code, or creating financial obligations. It is a pure read operation that queries existing data (arXiv papers) with no side effects. Low severity because reading academic papers presents minimal risk of misuse by an AI agent.
From the tool's definition Tool description states 'Read the full content of a stored paper in markdown format' - the verb 'Read' and the retrieval of content without modification are explicit.
Attacks that exploit this kind of access
Read the full content of a stored paper in markdown format. It is categorised as a Read tool in the ArXiv MCP Server MCP Server, which means it retrieves data without modifying state.
Register the ArXiv MCP Server MCP server in PolicyLayer and add a rule for read_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 ArXiv MCP Server. Nothing to install.
read_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 read_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 read_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.
read_paper is provided by the ArXiv MCP Server MCP server (wr-web/apr). 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 →