voice_transcribe
Transcribe short audio to text via OpenAI Whisper. BYOK ONLY (your own OpenAI key). Pass 'audio_url' (fetched SSRF-safely, ≤10MB) or 'audio_b64'. Sync — chunk long recordings. Returns {ok, transcript, chars}.
This record as markdown: /tools/com-wingmanprotocol-agent-gateway/voice-transcribe.md
What voice_transcribe does on Gateway
AI agents call voice_transcribe to retrieve information from Gateway without modifying anything. It is typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
| Parameter | Type | Required | Description |
|---|---|---|---|
handle | string | — | your handle (for BYOK via vault) |
prompt | string | — | context/vocabulary hint (optional) |
secret | string | — | your agent secret (for BYOK via vault) |
api_key | string | — | your OpenAI key (BYOK, inline) |
key_ref | string | — | vault entry holding your OpenAI key (BYOK alt) |
language | string | — | ISO code hint, e.g. en (optional) |
audio_b64 | string | — | base64 audio (alternative to audio_url) |
audio_url | string | — | http(s) URL of the audio (≤10MB) |
content_type | string | — | e.g. audio/mpeg (helps format detection) |
Parameters from the server's own tool schema.
Why voice_transcribe is rated Low
This is a pure data retrieval and conversion utility. It processes audio input and returns transcribed text. The BYOK (bring your own key) model, SSRF-safe fetching, and size limits (≤10MB) are security features that reduce risk. No reversible writes, code execution, destruction, or financial operations occur.
From the tool's definition Tool 'transcribes short audio to text' and 'returns {ok, transcript, chars}' — retrieves/converts audio input to text data with no side effects, modifications, or deletions. No data is created, modified, executed, or destroyed by this operation.
Risk signalsHandles credentials or secrets (api_key)
Attacks that exploit this kind of access
The rule that runs voice_transcribe safely
PolicyLayer is an MCP gateway: it sits between your AI agents and Gateway, and checks every tool call against a rule you set before the call runs. Nothing changes on the server itself. For voice_transcribe, this is the rule to start with:
voice_transcribe is read-only, so it stays allowed. Everything else on the server is denied unless you say otherwise.
The button opens the PolicyLayer dashboard: create your workspace, connect Gateway, apply this rule, and every voice_transcribe call is checked against it from then on.
Questions about voice_transcribe
Transcribe short audio to text via OpenAI Whisper. BYOK ONLY (your own OpenAI key). Pass 'audio_url' (fetched SSRF-safely, ≤10MB) or 'audio_b64'. Sync — chunk long recordings. Returns {ok, transcript, chars}. It is categorised as a Read tool in the Gateway MCP Server, which means it retrieves data without modifying state.
voice_transcribe accepts 9 parameters: handle, prompt, secret, api_key, key_ref, language, audio_b64, audio_url, content_type. The full parameter table on this page comes from the server's own tool schema.
Register the Gateway MCP server in PolicyLayer and add a rule for voice_transcribe: 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 Gateway. Nothing to install.
voice_transcribe 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 voice_transcribe 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 voice_transcribe. 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.
voice_transcribe is provided by the Gateway MCP server (https://wingmanprotocol.com/mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
More on Gateway, and thousands of servers like it.
This server
Across the catalogue