AI agents use connect_to_server to create or update resources in Pl — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Pl environment.
This tool performs a state-modifying operation by establishing a connection to an external server, which persists configuration or session state. While non-destructive and not executing arbitrary code, it creates a new system state. This is most appropriately classified as Write rather than Read (which only retrieves data) or Execute (which triggers code/operations with side effects beyond connection management).
From the tool's definition "Connect to a Platforma server" indicates creating or establishing a connection, which modifies the client state and configuration. This is a reversible write operation (the connection can be disconnected via the sibling 'disconnect' tool).
Attacks that exploit this kind of access
Connect to a Platforma server. Use list_connections to see saved servers. It is categorised as a Write tool in the Pl MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Pl MCP server in PolicyLayer and add a rule for connect_to_server: 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 Pl. Nothing to install.
connect_to_server 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 connect_to_server 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 connect_to_server. 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.
connect_to_server is provided by the Pl MCP server (@milaboratories/pl-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 →