AI agents call what_do_i_do to retrieve information from SEED MCP without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves and returns instructions or guidance for conversation behavior. It performs a simple query/lookup operation with no side effects, data modification, code execution, deletion, or financial impact. The verb 'Get' combined with the passive nature of retrieving static instructions classifies this as a Read operation with low severity due to minimal blast radius if misused.
From the tool's definition Tool name 'what_do_i_do' and description 'Get master instructions for what to do in conversations' indicate a retrieval operation that fetches instructional data without modification or external effects.
Attacks that exploit this kind of access
Get master instructions for what to do in conversations. It is categorised as a Read tool in the SEED MCP MCP Server, which means it retrieves data without modifying state.
Register the SEED MCP server in PolicyLayer and add a rule for what_do_i_do: 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 SEED MCP. Nothing to install.
what_do_i_do 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 what_do_i_do 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 what_do_i_do. 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.
what_do_i_do is provided by the SEED MCP server (sancovp/seed-mcp). 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 →