AI agents call lsp_hover to retrieve information from Lsp without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
lsp_hover retrieves type and documentation metadata from a language server without modifying, executing, or deleting any code or data. It is a read-only query operation that provides introspective information about code symbols. The context (LSP server for code understanding) and the explicit 'without reading its implementation' language confirm this is passive analysis only.
From the tool's definition Tool description explicitly states 'Get the type signature and documentation for any symbol' and 'understand what a function accepts/returns or what type a variable has, without reading its implementation.' These are pure information retrieval operations with…
Attacks that exploit this kind of access
Get the type signature and documentation for any symbol. Use to understand what a function accepts/returns or what type a variable has, without reading its implementation. It is categorised as a Read tool in the Lsp MCP Server, which means it retrieves data without modifying state.
Register the Lsp MCP server in PolicyLayer and add a rule for lsp_hover: 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 Lsp. Nothing to install.
lsp_hover 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 lsp_hover 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 lsp_hover. 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.
lsp_hover is provided by the Lsp MCP server (lsp-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 →