Sherlock Calls
for Vapi
Every Vapi conversation is a case file — transcripts, tool calls, latency events, cost records. Sherlock reads all of it. Invite-only beta: be among the first teams to run natural language investigations across your entire Vapi agent fleet from Slack.
TL;DR — What beta users get access to
- 1
Sherlock Calls for Vapi is in invite-only beta — connecting to your Vapi account via API key, giving your team natural language access to every conversation from Slack. Beta users get dedicated onboarding.
- 2
Beta users can ask Sherlock anything about their Vapi agents: 'Which calls had the highest latency this week?' 'Why did this tool invocation fail?' — answered with sourced data in under 5 seconds.
- 3
Sherlock correlates Vapi conversation data with your telephony provider and CRM to give you the full call lifecycle in a single query. Apply now to be among the first teams to access this.
350,000+
Developers & companies using Vapi
150M+
Calls processed on Vapi
0
Code to write for investigation
About Vapi
What is Vapi?
Vapi is the leading voice AI infrastructure platform for developers, with 350,000+ developers and companies — from startups to Fortune 500 — using it to build real-time voice AI agents. It has processed 150M+ calls and raised $20M in December 2024. Vapi provides a unified API layer over major LLMs, TTS providers, and telephony — powering outbound sales, inbound support, and scheduling agents at enterprise scale.
About Sherlock Calls
What is Sherlock Calls?
Sherlock Calls is an AI call detective that lives in your Slack workspace. It connects to voice providers, CRMs, and ad platforms, then answers any question about your calls in plain English — in under 5 seconds, with no code, no new dashboards, and no SDK required. Think of it as a senior analyst always on call, inside the tools your team already uses.
Beta Setup
Connect Vapi to Sherlock in 2 minutes
No code, no webhooks, no new dashboards. Beta users get direct onboarding support.
- 1
Connect Vapi via API key
Add your Vapi API key in the Sherlock Calls dashboard — read-only access to all conversations, agents, and cost data. No code changes or webhook configuration required.
- 2
Add Sherlock to Slack
Install the Sherlock Calls Slack app to your workspace in one click. Your team can ask questions from any channel immediately.
- 3
Debug your first conversation
Ask 'Show me all calls where tool invocations failed this week' — Sherlock queries Vapi, correlates latency and cost data, and delivers a sourced answer in under 5 seconds.
Beta Use Cases
What beta teams are asking Sherlock about Vapi
- SC“Why is my Vapi agent taking over 3 seconds to respond on certain calls?”
- SC“Find all conversations where tool calls failed in the last 7 days”
- SC“What is the average cost per call for Agent X vs Agent Y this month?”
- SC“Show me the full conversation trace for call ID [X] including all tool calls”
- SC“Which calls had the highest latency this week and what caused it?”
Beta Features
What beta users can do with Vapi
Conversation Debugging
Trace through every step of a Vapi conversation. See tool calls, LLM responses, and TTS output in a unified timeline — from a single Slack question.
Latency Analysis
Identify where latency creeps in — LLM response time, TTS generation, tool call execution. Ask 'Which calls had over 2 second latency?' and get a ranked breakdown.
Cost Optimization
Break down costs per conversation across LLM, TTS, and telephony. Find which agents cost the most, which calls cost the most, and why.
Agent Comparison
Compare performance across different agent configurations. Get data-driven answers on which prompts, models, and voices perform best.
Error Tracking
Find conversations that errored out. Trace the full error chain to identify whether the issue is in your tools, prompts, or Vapi itself.
Cross-Provider Correlation
Correlate Vapi conversations with your telephony provider (Twilio, Vonage) to get the complete call lifecycle — from SIP event to final transcript.
AI Tools
4 specialized tools, zero configuration
Sherlock picks the right tool automatically. You just ask in plain English — the AI knows which APIs to call and how to correlate the results.
Beta users get direct access to all tools below — with personal onboarding from the Sherlock team.
FAQ
Common questions about Sherlock + Vapi
How does Sherlock Calls connect to Vapi?
- Sherlock Calls connects to your Vapi account via API key with read-only access. No webhooks, no code changes, and no Vapi dashboard configuration is required. Setup takes under 2 minutes.
Can Sherlock Calls analyze Vapi conversation latency?
- Yes. Sherlock Calls can analyze end-to-end latency for any Vapi conversation — breaking down LLM response time, TTS generation time, and tool call execution time. Ask 'Which calls had latency over 2 seconds last week?' for an immediate ranked answer.
Can Sherlock Calls access Vapi conversation transcripts?
- Yes. Sherlock Calls can retrieve and search across Vapi conversation transcripts. Ask 'Find conversations where the agent said it couldn't help' or 'Show me the full transcript for this call ID' — both return sourced results.
Can Sherlock correlate Vapi data with Twilio?
- Yes. Sherlock builds unified timelines correlating Vapi conversation data with your underlying telephony provider — Twilio, Vonage, or others. This lets you trace the full call lifecycle from SIP handoff to final transcript.
What Vapi data can Sherlock Calls access?
- Sherlock Calls can access Vapi conversations, transcripts, tool call records, latency data, cost breakdowns, and agent configurations. All data is queried in real time via the Vapi API — nothing is replicated or stored.
Apply for early access to Sherlock + Vapi
We're accepting a select group of beta users to shape the Vapi integration. Tell us about your stack and we'll reach out personally if you're a fit.
Even more powerful combined
Also works with