AI agents call get_recommendations to retrieve information from Semanticscholar without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool queries the Semantic Scholar API to fetch recommendations—a read-only operation that retrieves data without side effects. It does not create, modify, delete, or execute anything. It aligns with the Read category pattern of 'search, list, get, fetch' operations.
From the tool's definition The tool description states it 'retrieves recommended related papers based on a specified paper' (基于指定论文获取推荐的相关论文). This is purely a retrieval/query operation with no modification, deletion, or execution of external code.
Documented attack patterns abuse exactly the kind of access get_recommendations gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Semanticscholar, and nothing reaches the server without passing your rules. This is the rule we recommend for get_recommendations:
{
"version": "1",
"default": "deny",
"tools": {
"get_recommendations": {}
}
}get_recommendations 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.
基于指定论文获取推荐的相关论文. It is categorised as a Read tool in the Semanticscholar MCP Server, which means it retrieves data without modifying state.
Register the Semanticscholar MCP server in PolicyLayer and add a rule for get_recommendations: 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 Semanticscholar. Nothing to install.
get_recommendations 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 get_recommendations 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 get_recommendations. 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.
get_recommendations is provided by the Semanticscholar MCP server (xbghc/semanticscholar-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Semanticscholar, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.
Free to start. No card required.
9 Semanticscholar tools catalogued and risk-classified — across an index of 43,000+ MCP servers.