Proxies requests to a target agent using appropriate client libraries. Supports generate and stream interactions. Use server:agentId format for multi-server environments with agent name conflicts.
AI agents invoke callAgent to trigger actions in Mcp Agent Proxy. 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 executes requests against external agents (Mastra, LangGraph, etc.), triggering agent workflows whose effects depend entirely on the arguments passed and the agent's capabilities. The downstream agents could perform any action including destructive or financial operations, making this an Execute-category tool with high severity due to the unpredictable blast radius of chained agent calls.
From the tool's definition 'Proxies requests to a target agent', 'Supports generate and stream interactions', 'callAgent' - triggers external agent execution with arbitrary inputs
Documented attack patterns abuse exactly the kind of access callAgent gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Mcp Agent Proxy, and nothing reaches the server without passing your rules. This is the rule we recommend for callAgent:
{
"version": "1",
"default": "deny",
"tools": {
"callAgent": {
"limits": [
{
"counter": "callagent_rate",
"window": "minute",
"max": 10,
"scope": "grant"
}
]
}
}
}callAgent 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.
Free to start. No card required.
Proxies requests to a target agent using appropriate client libraries. Supports generate and stream interactions. Use server:agentId format for multi-server environments with agent name conflicts. It is categorised as a Execute tool in the Mcp Agent Proxy MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Mcp Agent Proxy MCP server in PolicyLayer and add a rule for callAgent: 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 Mcp Agent Proxy. Nothing to install.
callAgent 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 callAgent 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 callAgent. 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.
callAgent is provided by the Mcp Agent Proxy MCP server (mashh-lab/mcp-agent-proxy). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Mcp Agent Proxy, 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.
5 Mcp Agent Proxy tools catalogued and risk-classified — across an index of 43,000+ MCP servers.