https://retellai.com/?utm_source=discord&utm_medium=seo logo
e
a
Thanks for tagging me @enes (ekkovoice). I'll look through relevant sources and get back to you shortly.
### Answer I don't have specific information about Clinicminds in the available sources, so I cannot guarantee 100% compatibility with that particular application. However, Retell AI supports API integrations that allow voice agents to connect with external systems like CRMs, databases, and other tools. [(1) ]() The platform can exchange data and trigger actions automatically through APIs, enabling voice agents to retrieve data, update backend systems, and personalize responses in real time. Whether Retell AI works with Clinicminds would depend on: - Whether Clinicminds offers an API that Retell AI can connect to - The specific actions you need the voice agent to perform (booking appointments, retrieving patient info, etc.) Retell AI also offers pre-defined tool calling capabilities including appointment booking via Cal.com, and supports custom tools for more tailored actions. [(2) ]() To determine compatibility, I'd recommend: 1. Checking if Clinicminds has an API available 2. Testing with Retell AI's 60-minute free trial [(3) ]() 3. Reaching out to [Contact Sales]() for specific integration questions Would you like a general answer about how Retell AI's API integration capabilities work?
Feel free to tag @AI Evy with additional questions. I also appreciate feedback, it helps me improve.
e
@AI Evy I’m building a voice agent with Retell and want to connect it to Clinicminds for real-time appointment booking / rescheduling / cancelling. Clinicminds doesn’t offer a full open API yet, but exposes their online scheduler API: Auth via Bearer token + User-Agent GraphQL endpoint for fetching data (services, locations, times, UUIDs) JSON API for actions (book / verify via SMS / reschedule / cancel / refunds) Booking flow requires: Fetch services + availability (GraphQL) Create booking (JSON → sends SMS) Verify booking with code They warn the API may change and isn’t officially public/stable. My question: 👉 Is this kind of multi-step scheduler flow something Retell agents can reliably orchestrate in production? Including calling GraphQL + JSON endpoints, handling SMS verification steps, retries, and fallbacks?
a
Thanks for the follow-up, @enes (ekkovoice). I'll get back to you soon.
### Answer Yes, Retell AI can orchestrate multi-step workflows like the Clinicminds booking flow you described. Here's how the platform supports this: ## Custom Function Calling Retell agents support custom tools that can call external HTTP endpoints. [(1) ]() You can configure tools with: - Custom URLs (including your GraphQL and JSON endpoints) - HTTP methods (GET, POST, etc.) - Headers for Bearer token authentication
Copy code
javascript
{
  "type": "custom",
  "name": "get_customer_info",
  "description": "Get customer information from database",
  "tool_id": "tool_001",
  "url": "<https://api.example.com/customer%22,>
  "method": "GET"
}
int handling, but custom functions support POST requests to any URL, which should work for GraphQL queries For complex integrations like this, I'd recommend reaching out to [Contact Sales]() to discuss your specific requirements.
## Multi-Step Workflow Support Retell's Conversation Flow allows you to design multi-step processes with conditional logic. You can: - Break complex tasks into sequential nodes (fetch availability → create booking → verify code) - Use edges with transition conditions to move between steps - Store and reference data across steps using dynamic variables [(2) ]() For example, prompts can guide users through steps like: [(3) ]() 1. Confirm the service type 2. Collect time and date using dynamic variables like
{{preferred_date}}
3. Validate input and offer alternatives using
{{alternative_times}}
## Handling SMS Verification Retell agents can now send Outbound SMS directly to customers and handle two-way SMS conversations. [(4) ]() This means your agent could: - Trigger the booking creation (which sends SMS via Clinicminds) - Prompt the user for the verification code - Submit the code back to the JSON API ## MCP Integration for External APIs Retell supports *MCP (Model Context Protocol) * for connecting to external services. With MCP, your agents can: - Call any HTTP-based service - Trigger workflows and fetch live data in real time - Dynamically adapt the conversation based on structured responses ## Considerations for Production A few caveats for your specific use case: - **Unstable API warning**: Since Clinicminds warns their API may change, you'll need to monitor for breaking changes and update your custom functions accordingly - **Retries/fallbacks**: You can implement fallback logic within conversation flow nodes and use webhook timeout controls (configurable timeout settings are available) - **GraphQL specifics**: The sources don't explicitly document GraphQL endpo
Feel free to tag @AI Evy with additional questions. I also appreciate feedback, it helps me improve.
​ To learn more, see the sources I considered: [[1] Create Conversation Flow](https://docs.retellai.com/api-references/create-conversation-flow) [[2] Dynamic Variables](https://docs.retellai.com/build/dynamic-variables) [[3] 5 Useful Prompts for AI Agent Builders | Retell AI](https://www.retellai.com/blog/5-useful-prompts-for-building-ai-voice-agents-on-retell-ai) [[4] Platform Changelogs | Retell AI](https://www.retellai.com/changelog) ​