Hand off an in-flight task to a human operator with a full context bundle: transcript, prior actions, identifiers, and a recommended next step. EXAMPLE USER QUERIES THAT MATCH THIS TOOL: user: "I'm stuck — get a human at smb_xyz to call me back" -> call escalate_to_human({"smb_id": "smb_xyz", "re...
Part of the Agent Broker server.
Free to start. No card required.
AI agents call escalate_to_human to retrieve information from Agent Broker without modifying any data. This is common in research, monitoring, and reporting workflows where the agent needs context before taking action. Because read operations don't change state, they are generally safe to allow without restrictions -- but you may still want rate limits to control API costs.
Even though escalate_to_human only reads data, uncontrolled read access can leak sensitive information or rack up API costs. An agent caught in a retry loop could make thousands of calls per minute. A rate limit gives you a safety net without blocking legitimate use.
Read-only tools are safe to allow by default. No rate limit needed unless you want to control costs.
{
"version": "1",
"default": "deny",
"tools": {
"escalate_to_human": {}
}
}See the full Agent Broker policy for all 14 tools.
These attack patterns abuse exactly the kind of access escalate_to_human gives an agent. Each links to the full case and the policy that stops it:
Other read tools across the catalogue. The same approach applies to each: allow, with a rate cap to control cost.
Hand off an in-flight task to a human operator with a full context bundle: transcript, prior actions, identifiers, and a recommended next step. EXAMPLE USER QUERIES THAT MATCH THIS TOOL: user: "I'm stuck — get a human at smb_xyz to call me back" -> call escalate_to_human({"smb_id": "smb_xyz", "reason": "agent_blocked", "summary": "Cannot resolve via automated channels"}) WHEN TO USE: Use when automated resolution has failed after channel-fallback exhaustion, when the task requires human judgment, or when the customer has explicitly requested human contact. WHEN NOT TO USE: Do not use as a first resort. Escalate only after automated resolution attempts. COST: $0.2 per_escalation LATENCY: ~2000ms EXECUTION: async_by_default (use get_outcome to retrieve result). It is categorised as a Read tool in the Agent Broker MCP Server, which means it retrieves data without modifying state.
Register the Agent Broker MCP server in PolicyLayer and add a rule for escalate_to_human: 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 Agent Broker. Nothing to install.
escalate_to_human 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 escalate_to_human 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 escalate_to_human. 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.
escalate_to_human is provided by the Agent Broker MCP server (https://agent-broker-edge.basil-agent.workers.dev/mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Deterministic rules across all 14 Agent Broker tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.
Free to start. No card required.
4,600+ MCP servers and 31,000+ tools scanned and risk-classified.