Convert text to speech and save as an audio file. This will use the configured TTS engine to generate speech from text and save it to the specified output path. Args: text: The text to convert to speech (optional if text_file_path is provided) text_file_path: Path to a text file containing the te...
AI agents use narrate to create or update resources in Speech MCP — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Speech MCP environment.
The tool creates a new audio file at a specified location, which is a reversible write operation. It does not execute arbitrary code, delete files, move money, or retrieve sensitive data. The primary action is file creation. Severity is low because the blast radius of misuse (generating unwanted audio files) is minimal and the operation is easily reversible by deleting the output file.
From the tool's definition Tool description states it will 'Convert text to speech and save as an audio file' and 'save it to the specified output path.' This creates a new file on the filesystem.
Documented attack patterns abuse exactly the kind of access narrate gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Speech MCP, and nothing reaches the server without passing your rules. This is the rule we recommend for narrate:
{
"version": "1",
"default": "deny",
"tools": {
"narrate": {
"limits": [
{
"counter": "narrate_rate",
"window": "minute",
"max": 30,
"scope": "grant"
}
]
}
}
}narrate stays usable, but capped — an agent stuck in a loop can't make hundreds of changes a minute. Everything else on the server is denied unless you say otherwise.
Free to start. No card required.
Convert text to speech and save as an audio file. This will use the configured TTS engine to generate speech from text and save it to the specified output path. Args: text: The text to convert to speech (optional if text_file_path is provided) text_file_path: Path to a text file containing the text to narrate (optional if text is provided) output_path: Path where to save the audio file (.wav) Returns: A message indicating success or failure of the operation. It is categorised as a Write tool in the Speech MCP MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Speech MCP server in PolicyLayer and add a rule for narrate: 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 Speech MCP. Nothing to install.
narrate is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.
Yes. Add a rate_limit block to the narrate 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 narrate. 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.
narrate is provided by the Speech MCP server (kvadratni/speech-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Speech MCP, 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.
7 Speech MCP tools catalogued and risk-classified — across an index of 43,000+ MCP servers.