AI agents use inferReturnType to create or update resources in Ts — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Ts environment.
An AI agent can call inferReturnType faster than any human can review — one bad instruction and it creates or modifies resources in Ts by the hundred, each call as confident as the last.
Attacks that exploit this kind of access
Add an explicit return type annotation to a function, inferred by TypeScript. Position must be on the function name or declaration keyword. It is categorised as a Write tool in the Ts MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Ts MCP server in PolicyLayer and add a rule for inferReturnType: 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 Ts. Nothing to install.
inferReturnType is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.
Yes. Add a rate_limit block to the inferReturnType 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 inferReturnType. 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.
inferReturnType is provided by the Ts MCP server (ts-mcp-server). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.