AI agents invoke rpc_reconnect to trigger actions in Rpcclient. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call — builds kicked off, notifications sent, workflows started.
This tool performs an active operation on an existing session — reconnecting transport and clearing stale handles. While not destructive or financial, it triggers external operations (re-establishing RPC connection, clearing handles) whose effects depend on the session state. Misuse could disrupt an active automation session or clear valid handles unexpectedly.
From the tool's definition Reconnect an existing session to recover from broken transport; clears stale handles.
Attacks that exploit this kind of access
Reconnect an existing session to recover from broken transport; clears stale handles. It is categorised as a Execute tool in the Rpcclient MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Rpcclient MCP server in PolicyLayer and add a rule for rpc_reconnect: 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 Rpcclient. Nothing to install.
rpc_reconnect is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.
Yes. Add a rate_limit block to the rpc_reconnect 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 rpc_reconnect. 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.
rpc_reconnect is provided by the Rpcclient MCP server (appknox/rpcclient-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 →