AI agents call list_recipes to retrieve information from Mcp Docs without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves and enumerates documentation/recipes for implementation guidance. It performs a read-only query operation that returns reference material. No data is created, modified, deleted, or executed. The blast radius of misuse is minimal—an agent could only retrieve already-public documentation content.
From the tool's definition Tool name is 'list_recipes' and description states it 'List[s] curated end-to-end implementation recipes' — a listing/retrieval operation with no side effects.
Attacks that exploit this kind of access
List curated end-to-end implementation recipes for common ProsodyAI integration tasks (e.g. add prosody to a LiveKit agent, stream from a browser, wire the LangChain tool, define KPIs). It is categorised as a Read tool in the Mcp Docs MCP Server, which means it retrieves data without modifying state.
Register the Mcp Docs MCP server in PolicyLayer and add a rule for list_recipes: 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 Mcp Docs. Nothing to install.
list_recipes 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 list_recipes 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 list_recipes. 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.
list_recipes is provided by the Mcp Docs MCP server (prosodyai/mcp-docs). 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 →