This record as markdown: /tools/io-github-heznpc-iconnect/ai-agent.md
What ai_agent does on Iconnect
AI agents invoke ai_agent to trigger actions in Iconnect. 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.
Why ai_agent is rated High
The 'ai_agent' tool executes arbitrary prompts through an Apple AI system. While the description is somewhat terse, the combination of 'Run' (verb indicating execution) and 'prompt' (variable user input) classifies this as Execute rather than Read.
From the tool's definition Tool name 'ai_agent' combined with description 'Run a prompt through Apple' indicates execution of an AI agent with arbitrary prompts. This is an Execute operation that runs code/models whose effects depend on the prompt argument.
Attacks that exploit this kind of access
The rule that runs ai_agent safely
PolicyLayer is an MCP gateway: it sits between your AI agents and Iconnect, and checks every tool call against a rule you set before the call runs. Nothing changes on the server itself. For ai_agent, this is the rule to start with:
ai_agent stays usable, but rate-capped: a runaway agent can't fire it dozens of times a minute. Everything else on the server is denied unless you say otherwise.
The button opens the PolicyLayer dashboard: create your workspace, connect Iconnect, apply this rule, and every ai_agent call is checked against it from then on.
Questions about ai_agent
Run a prompt through Apple. It is categorised as a Execute tool in the Iconnect MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Iconnect MCP server in PolicyLayer and add a rule for ai_agent: 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 Iconnect. Nothing to install.
ai_agent 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 ai_agent 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 ai_agent. 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.
ai_agent is provided by the Iconnect MCP server (iconnect-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
More on Iconnect, and thousands of servers like it.
Across the catalogue