Transcribes audio via Whisper. Preferred: audio_url (most token-efficient; server fetches bytes). audio_base64 is for small clips only (<= ~60KB raw per call). audio_path only works when the MCP host shares a filesystem with the caller (often false on Claude.ai / Claude Code). For larger payloads...
AI agents invoke transcribe-audio to trigger actions in Video Toolkit. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call — builds kicked off, notifications sent, workflows started.
This tool triggers an external AI speech-to-text operation (Whisper), fetches remote audio data, re-encodes it, and runs inference — all server-side operations with external dependencies. It is not a simple read/query; it executes a multi-step processing pipeline. Severity is medium since misuse could involve fetching arbitrary URLs or processing unintended audio, but there is no financial or destructive impact.
From the tool's definition Transcribes audio via Whisper... server fetches bytes... Server re-encodes to Opus 16kHz mono 16kbps before Whisper
Attacks that exploit this kind of access
Transcribes audio via Whisper. Preferred: audio_url (most token-efficient; server fetches bytes). audio_base64 is for small clips only (<= ~60KB raw per call). audio_path only works when the MCP host shares a filesystem with the caller (often false on Claude.ai / Claude Code). For larger payloads in sandboxed environments, use transcribe_upload_start / transcribe_upload_append / transcribe_upload_finalize. Server re-encodes to Opus 16kHz mono 16kbps before Whisper unless skip_compression=true. Long audio (>5min) or async=true returns a job_id; poll transcribe_get_job. It is categorised as a Execute tool in the Video Toolkit MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Video Toolkit MCP server in PolicyLayer and add a rule for transcribe-audio: 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 Video Toolkit. Nothing to install.
transcribe-audio is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.
Yes. Add a rate_limit block to the transcribe-audio 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 transcribe-audio. 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.
transcribe-audio is provided by the Video Toolkit MCP server (jamesanz/transcript-mcp). 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 →