AI agents call lean_references to retrieve information from Lean LSP MCP without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
References lookups in language servers are fundamentally read-only queries that retrieve metadata about code structure without modifying, executing, or deleting anything. No blast radius for misuse beyond potentially excessive queries. Low severity due to benign nature of the operation.
From the tool's definition Tool name 'lean_references' suggests querying or retrieving references (cross-references, usages, or definitions) within a Lean codebase. Context shows this is part of a Lean LSP server providing analysis and documentation tools.
Documented attack patterns abuse exactly the kind of access lean_references gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Lean LSP MCP, and nothing reaches the server without passing your rules. This is the rule we recommend for lean_references:
{
"version": "1",
"default": "deny",
"tools": {
"lean_references": {}
}
}lean_references 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.
lean_references. It is categorised as a Read tool in the Lean LSP MCP MCP Server, which means it retrieves data without modifying state.
Register the Lean LSP MCP server in PolicyLayer and add a rule for lean_references: 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 Lean LSP MCP. Nothing to install.
lean_references 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 lean_references 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 lean_references. 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.
lean_references is provided by the Lean LSP MCP server (ooo0ooo/lean-lsp-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Deterministic rules across all 23 Lean LSP MCP tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.
Free to start. No card required.
23 Lean LSP MCP tools catalogued and risk-classified — across an index of 42,500+ MCP servers.