AI agents call request_human_approval as a supporting operation in Z Zero workflows.
| Parameter | Type | Required | Description |
|---|---|---|---|
situation | string | Yes | Clear description of what the bot found, e.g. 'Checkout shows $20 total (includes $3 tax) but current token is only $15' |
current_token | string | — | Current active token ID if any |
alternative_action | string | — | Alternative option if available |
recommended_action | string | Yes | What the bot recommends doing, e.g. 'Cancel current $15 token and issue a new $20 token' |
Parameters from the server's own tool schema.
This tool does not itself read, write, execute, delete, or move money — it pauses execution and solicits human approval. It is a control/gate mechanism rather than an action with side effects. However, it lives on a financial server and gates financial operations, so misuse (e.g., silently bypassing or spoofing approval) could have downstream financial impact; hence medium-low concern.
From the tool's definition Pause and ask the user for approval before risky actions (price mismatch, large amount, unusual request)
Attacks that exploit this kind of access
Pause and ask the user for approval before risky actions (price mismatch, large amount, unusual request). It is categorised as a Other tool in the Z Zero MCP Server, which means it performs auxiliary operations.
request_human_approval accepts 4 parameters: situation, current_token, alternative_action, recommended_action. Required: situation, recommended_action. The full parameter table on this page comes from the server's own tool schema.
Register the Z Zero MCP server in PolicyLayer and add a rule for request_human_approval: 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 Z Zero. Nothing to install.
request_human_approval is a Other tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the request_human_approval 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 request_human_approval. 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.
request_human_approval is provided by the Z Zero MCP server (z-zero-mcp-server). 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 →