AI agents call get_agent to retrieve information from Pantheon without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The tool retrieves or queries data (a specific agent by name) without modifying, deleting, or executing anything. It is a straightforward read operation similar to a GET request or fetch call. The low severity reflects that accessing agent instructions poses minimal risk even if misused by an AI agent, as it only retrieves information already available on the server.
From the tool's definition Tool name is 'get_agent' and description states 'Get a specific agent by name' - this is a retrieval operation with no side effects.
Documented attack patterns abuse exactly the kind of access get_agent gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Pantheon, and nothing reaches the server without passing your rules. This is the rule we recommend for get_agent:
{
"version": "1",
"default": "deny",
"tools": {
"get_agent": {}
}
}get_agent is read-only, so it stays allowed — but everything else on the server is denied unless you say otherwise.
Free to start. No card required.
Get a specific agent by name. It is categorised as a Read tool in the Pantheon MCP Server, which means it retrieves data without modifying state.
Register the Pantheon MCP server in PolicyLayer and add a rule for get_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 Pantheon. Nothing to install.
get_agent 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 get_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 get_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.
get_agent is provided by the Pantheon MCP server (valado/pantheon-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Pantheon, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.
Free to start. No card required.
3 Pantheon tools catalogued and risk-classified — across an index of 43,000+ MCP servers.