canonical: https://jentic.com/apis/swaggerhub.feedyou/feedyou-bot-telephony-api

# Feedyou Bot Telephony API

This API allows to use Feedyou bot with telephony PBX which implements TTS/STT voice processing on its own. - When user calls PBX, `call_start` event is sent to the bot - Bot loads set of intro messages and sends them one by one using `queue_tts` to the PBX (endpoint where PBX is listening for events from bot should be defined in `callback_endpoint` property) - PBX should transform them to the voi. The API exposes 8 endpoints.

## For AI agents

Programmatically notify bot about new incoming call, pass text recognized by stt to the bot. Covers 8 operations.

## Scope

Does not handle payments, crm, or developer tools - use for communications only.

## Capabilities

- Notify bot about new incoming call
- Pass text recognized by STT to the bot
- Listen timeout has been reached
- Play message using TTS engine to pending call, optionally switching call to requ
- Get list of calls for specified period of time
- Download recording file

## Use cases

### Communications Operations

Use the Feedyou Bot Telephony API to perform communications operations programmatically. The API provides 8 endpoints covering core functionality including notify bot about new incoming call, pass text recognized by stt to the bot, listen timeout has been reached.

Example prompt: Call POST /call_start to notify bot about new incoming call

### Automated Call_End Management

Automate call_end operations by combining multiple Feedyou Bot Telephony API endpoints. Agents can pass text recognized by stt to the bot and then listen timeout has been reached in a single workflow.

Example prompt: Call POST /recognized_text to pass text recognized by stt to the bot, then verify the result

### AI Agent Integration via Jentic

AI agents discover and call Feedyou Bot Telephony API endpoints through Jentic without managing credentials directly. An agent searches for the required operation by intent, receives the matching endpoint schema, and executes the call with Jentic-managed authentication. This eliminates the need to read API documentation or handle none tokens manually.

Example prompt: Search Jentic for 'notify bot about new incoming call', load the operation schema, and execute with Jentic-managed credentials

## Key endpoints

| Method | Path | Description |
| --- | --- | --- |
| POST | `/call_start` | Notify bot about new incoming call |
| POST | `/recognized_text` | Pass text recognized by STT to the bot |
| POST | `/listen_timeout_reached` | Listen timeout has been reached |
| POST | `/queue_tts` | Play message using TTS engine to pending call, optionally switching call to requ |
| POST | `/call_end` | Notify bot about finished call |
| GET | `/get_calls_overview` | Get list of calls for specified period of time |
| GET | `/get_rec_file` | Download recording file |
| POST | `/get_call_debug` | Get call debug log |

## Key resources

- **Call_End** — Operations for call_end
- **Call_Start** — Operations for call_start
- **Get_Call_Debug** — Operations for get_call_debug
- **Get_Calls_Overview** — Operations for get_calls_overview
- **Get_Rec_File** — Operations for get_rec_file

## Why Jentic

- **Setup:** Wiring the Feedyou Bot Telephony API by hand means resolving your per-bot Azure host and coordinating the call lifecycle yourself across call start, recognized text, timeout, queued speech and call end. Through Jentic you install once, import Feedyou Bot Telephony API from the API Directory, store the connection once, and your agent calls it.
- **Permission scoping:** This API carries the call target in the request body, so scope the agent to the operations it needs, such as starting a call or fetching the calls overview. You choose the operations it may call, so ones like ending a call or queuing text-to-speech are not included unless you add them.
- **Credential handling:** This API declares no authentication, so there is no secret to hand around; any connection settings are held once by your own Jentic One instance and injected at execution time. They never enter the agent's prompt, logs, or context.
- **Discovery method:** Agents search Jentic by intent such as 'notify the bot about an incoming call' or 'get the recording file', and Jentic returns the matching Feedyou Bot Telephony API operation with its input schema so the agent calls the right endpoint without browsing the reference docs.

## Related APIs

- **Twilio** — Alternative communications API
- **Sendgrid** — Alternative communications API
- **Pusher** — Complementary communications API

## FAQ

### What authentication does the Feedyou Bot Telephony API use?

The Feedyou Bot Telephony API uses no authentication. Through Jentic, these credentials are stored encrypted in your Jentic One instance and injected at execution time, so raw secrets never enter the agent context.

### Can I notify bot about new incoming call with the Feedyou Bot Telephony API?

Yes. Use the POST /call_start endpoint. The API returns structured JSON responses that agents can parse and act on directly.

### What are the rate limits for the Feedyou Bot Telephony API?

Rate limits are not specified in the OpenAPI spec. Check the vendor documentation for current limits. Through Jentic, rate limiting is handled automatically with retry logic built into the execution layer.

### How do I notify bot about new incoming call through Jentic?

Install the Jentic SDK with pip install jentic, authenticate through Jentic One, the self-hosted execution layer, then search for 'notify bot about new incoming call'. Jentic returns the matching Feedyou Bot Telephony API operation with its input schema. Load the schema and execute the call - credentials are injected automatically.

### How many endpoints does the Feedyou Bot Telephony API have?

The Feedyou Bot Telephony API exposes 8 endpoints covering call_end, call_start, get_call_debug operations.

### Can I limit what my agent is allowed to do with the Feedyou Bot Telephony API?

Yes. Because you run Jentic One yourself, your own rules decide which of the Feedyou Bot Telephony API operations the agent may call, so you can allow only what a task needs, such as POST /call_start to notify the bot of an incoming call or GET /get_calls_overview to fetch the calls list. Operations like POST /call_end or POST /queue_tts stay unavailable to the agent unless you explicitly grant them. Any connection settings are held by your own Jentic One instance and injected at execution time rather than exposed to the agent.