Loading ToolBench
Voice Agent Factory MCP — deploy, eval, and audit phone agents built on Vapi. Differs from the official Vapi MCP: vertical templates, self-testing agents, audio forensics, prompt versioning, CI regression gates.
Download the 3 recording tracks of a call (mono/customer/assistant), run RMS analysis, and locate the noise source. Automates the manual audio investigation.
Deploy a full production phone agent for a market (fr, us, sv) and a vertical. Available verticals — fr: elagueur, plombier, dentiste, veterinaire. us: plumber, tree-service, dentiste, veterinaire. sv: plombier, dentiste, veterinaire. Creates the Vapi assistant with prompt, voice, transcriber and keywords baked in. One call = one working agent.
List the Vapi assistants on this account: name, id, model, voice, transcriber, creation date. Start here to find the assistantId for run_eval / prompt_diff / regression_gate.
Version and diff assistant prompts. snapshot = capture current prompt from Vapi into local history. diff = compare two versions. rollback = push an old prompt back.
CI for prompts. Runs run_eval against the current assistant and blocks the update if the score regressed vs the stored baseline. Returns allow/deny.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 11 | 2026-07-28+ | v2 |
The agent tests itself. Simulates N scripted scenarios via the LLM (no audio cost), scores behavior (close speed, no price, spelled confirmation, urgency handling), returns a /100 report with PASS/WARN/FAIL.
Have an agent place a real outbound phone call. Uses the first phone number on the Vapi account as caller ID. The customer number accepts E.164 (+14155550132) or local formats (10-digit US, 10-digit FR starting with 0, 8-digit SV).