👋 Hey Vapi Team, We’re running into a problem where our AI receptionist is not speaking the tool response after a successful get_pricing call — even though everything appears to be set up correctly. ✅ What’s Working Tool returns a valid response: json Copy Edit { "match_found": "true", "price_string": "$150.00" } request-complete block is defined: json Copy Edit { "type": "request-complete", "content": "The price is [price_string]. Would you like to go ahead and schedule the appointment?", "conditions": [ { "param": "match_found", "value": "true", "operator": "eq" } ] } Assistant prompt includes: "Do not speak immediately after tool calls; let the tool response speak instead." Tool is blocking, not async. ❌ What’s Going Wrong Despite everything above, the assistant still speaks manually after calling the tool: “Let me check the pricing for you. 1 moment, please.” This seems to override or suppress the request-complete message — even though the tool matched successfully. 🧪 Test Info Tool ID: eefd335d-9ba9-44db-9193-b5010f758543 Call ID: 1a7dfa3f-cac7-4069-9e66-eaf9a36ee725 Tool Call ID: call_DQ7c1KRVOZrUzBMRKoTL6EMQ Assistant ID: 8de39e0c-9a38-4350-9475-036d9e91f178 Timestamp: 2025-05-14 18:10:29 UTC Same issue happens across: Local Flask API Render-hosted version of the same endpoint 🙏 Request for Help Is this a known issue with request-complete getting suppressed by post-tool assistant speech? Should we tweak anything in the tool definition or prompt to guarantee tool response is spoken? Can you confirm if request-complete was queued but not spoken in this case? Thanks for helping us debug this last-mile issue! 🔧 Best, Adrian Torres 817-586-9634