Search PHANTOM VOICES voice clones by keyword (e.g.
AI agents call search_voices to retrieve information from Phantom Voices without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool performs a search/query operation to retrieve voice clone information from a database or service. It has no side effects, does not execute code or commands, does not modify data, and does not involve financial transactions. It is purely a retrieval operation matching the 'Read' category definition.
From the tool's definition Tool name 'search_voices' and description 'Search PHANTOM VOICES voice clones by keyword' indicate a read-only query operation that retrieves voice configuration data without modifying or executing anything.
Attacks that exploit this kind of access
Search PHANTOM VOICES voice clones by keyword (e.g. It is categorised as a Read tool in the Phantom Voices MCP Server, which means it retrieves data without modifying state.
Register the Phantom Voices MCP server in PolicyLayer and add a rule for search_voices: 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 Phantom Voices. Nothing to install.
search_voices 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 search_voices 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 search_voices. 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.
search_voices is provided by the Phantom Voices MCP server (todiwalaventures/phantom-voices-api). 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 →