Quick Replies Overview

Configure standardized message snippets and slash command shortcuts for agent chat acceleration

Quick Replies API

The Quick Replies API allows teams to define canned response templates and slash shortcuts (e.g. /greeting, /pricing, /refund-policy). During live WhatsApp customer chat sessions in Sayvy AI Inbox, human agents and conversational copilots can type shortcuts to instantly expand comprehensive, pre-approved messaging copy.

+-------------------------------------------------------------+
|              Quick Reply Shortcut Architecture              |
|                                                             |
|   +-----------------------+     +-----------------------+   |
|   | Human Agent in Inbox  |     | AI Copilot Assistant  |   |
|   |  Types: "/refund"     |     |   Suggests Answers    |   |
|   +-----------+-----------+     +-----------+-----------+   |
|               \                             /               |
|                v                           v                |
|              +-------------------------------+              |
|              |     Redis Low-Latency Cache   |              |
|              |     (300s TTL per Org Cache)  |              |
|              +---------------+---------------+              |
|                              |                              |
|                      (Cache Miss Fallback)                  |
|                              v                              |
|              +-------------------------------+              |
|              |   PostgreSQL Persistent Store |              |
|              |  (Indexed Organization Model) |              |
|              +---------------+---------------+              |
|                              |                              |
|                              v                              |
|              +-------------------------------+              |
|              | Instant Expansion in Composer |              |
|              | "We offer full 30-day refunds"|              |
|              +-------------------------------+              |
+-------------------------------------------------------------+

---

Key Capabilities

Instant Keyboard Expansion

Accelerate support response times with pre-configured shortcut triggers like /hello, /hours, or /bank-details.

High-Speed Redis Caching

Responses are served from Redis memory cache with automatic cache busting whenever snippets are created or deleted.

Standardized Messaging Quality

Ensure organizational consistency and compliance across every agent response without spelling or factual errors.

Omnichannel Reuse

Use quick reply templates seamlessly across WhatsApp, web widgets, and live ticketing channels.

---

Cache Invalidation Strategy

The Quick Replies system utilizes Redis caching for blazing-fast typeahead lookups in the UI:

• Listing Requests: Cached under quick_replies:{organization_id}:{limit}:{page} with a 300-second (5 minute) TTL.
• Eviction Triggers: Any call to POST /api/v1/quick-replies or DELETE /api/v1/quick-replies/{id} immediately purges all keys matching quick_replies:{organization_id}:* to prevent stale snippet data.

---

Available Endpoints

Method	Endpoint	Description
POST	/api/v1/quick-replies	Create a new quick reply snippet and invalidate cache
GET	/api/v1/quick-replies	Retrieve paginated quick replies from cache or database
DELETE	/api/v1/quick-replies/{quick_reply_id}	Permanently delete a quick reply snippet and clear cache