AI agents call key_info to retrieve information from Litellm without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves metadata and spend information associated with an API key. It queries existing data and returns information only—no creation, modification, deletion, or execution of operations. The action is read-only with minimal blast radius if misused; an agent could at most enumerate key details or track spending, but cannot invoke financial transactions or modify keys themselves.
From the tool's definition Tool name 'key_info' and description 'Get details and spend info for a LiteLLM Proxy API key' indicate retrieval of existing information without modification or side effects.
Attacks that exploit this kind of access
Get details and spend info for a LiteLLM Proxy API key (/key/info). It is categorised as a Read tool in the Litellm MCP Server, which means it retrieves data without modifying state.
Register the Litellm MCP server in PolicyLayer and add a rule for key_info: 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 Litellm. Nothing to install.
key_info 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 key_info 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 key_info. 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.
key_info is provided by the Litellm MCP server (litellm-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 →