AI agents call luogu_fetch_problem to retrieve information from Luogu without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves static problem data without any side effects. It performs a simple query operation to obtain problem information indexed by problem ID (pid). This is a classic Read operation with minimal risk—the worst outcome of misuse would be excessive API calls or data exfiltration of already-public problem statements.
From the tool's definition The tool description explicitly states it 'Fetch[es] a Luogu problem statement' and related metadata (formats, samples, tags, difficulty, source URL). The server is described as 'read-only' and the tool name contains 'fetch', a retrieval verb.
Attacks that exploit this kind of access
Fetch a Luogu problem statement, formats, samples, tags, difficulty, and source URL by pid. It is categorised as a Read tool in the Luogu MCP Server, which means it retrieves data without modifying state.
Register the Luogu MCP server in PolicyLayer and add a rule for luogu_fetch_problem: 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 Luogu. Nothing to install.
luogu_fetch_problem 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 luogu_fetch_problem 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 luogu_fetch_problem. 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.
luogu_fetch_problem is provided by the Luogu MCP server (kaiserunix/luogu-mcp-server). 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 →