Get papers similar to a given paper using Semantic Scholar
AI agents call library_recommend to retrieve information from Alexandria MCP without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool queries and retrieves similar papers based on semantic matching. It performs a lookup/search operation against existing data without creating, modifying, deleting, or executing anything. The severity is low because retrieving academic paper recommendations poses minimal risk even if misused by an agent.
From the tool's definition Tool description states 'Get papers similar to a given paper' — a retrieval operation with no side effects. The server context confirms it 'enables users to search and retrieve information' from digital libraries.
Attacks that exploit this kind of access
Get papers similar to a given paper using Semantic Scholar. It is categorised as a Read tool in the Alexandria MCP MCP Server, which means it retrieves data without modifying state.
Register the Alexandria MCP server in PolicyLayer and add a rule for library_recommend: 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 Alexandria MCP. Nothing to install.
library_recommend 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 library_recommend 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 library_recommend. 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.
library_recommend is provided by the Alexandria MCP server (the-40-thieves/alexandria-mcp). 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 →