Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.bland.ai/llms.txt

Use this file to discover all available pages before exploring further.

​
Introduction

Bland Messaging lets you run conversational agents over SMS, RCS, and iMessage from one place. The same pathways, personas, tools, memory, and analytics power every channel, so the flow you build for voice can drive your text conversations without rebuilding it.
Enterprise feature — Messaging is available on Enterprise plans. Contact your Account Executive or reach out to sales to get set up.

Pathway-driven replies

Use the same pathways and personas you already use for voice. Backchanneling and pause hints are stripped automatically so messages stay clean.

Cross-channel memory

A contact reached over SMS, iMessage, or voice keeps one shared history. Facts, summaries, and open items carry across channels.

Outcomes and citations

The same outcomes, citation schemas, and summary prompts you use on voice run on every messaging conversation.

One API surface

Send and receive across all three channels with the same /v1/sms/* endpoints. Channel is configured per number or selected per request.

​
Channels

Each channel has different reach and capabilities. You can use one number for one channel, or layer them so the same contact escalates to richer messaging when supported.

iMessage

Branded contact on Apple devices with photos, videos, voice memos, tapbacks, text effects, and end-to-end encryption.

RCS

Verified business profile, suggested replies, action buttons, cards, and media on supported Android devices and carriers.

SMS

Universal reach in the US. Plain text. Requires A2P registration.
If your number is provisioned for RCS or iMessage, Bland uses the richest channel the recipient supports and falls back when needed.

​
Attaching a pathway or persona

Pathways and personas attach to messaging numbers the same way they attach to phone numbers for calls. From the Messaging dashboard, pick a number and configure it with a pathway or a persona. The number then handles inbound messages using the same logic, tools, and memory you use on voice. If you already have a voice pathway, you can clone it for messaging and use the Sync action to keep both in sync as you iterate.

​
Memory across conversations

Memory can be enabled on any messaging channel and is shared with your voice agents. When memory is on, the same contact is recognized across SMS, RCS, iMessage, and voice, with one coherent history of facts, summaries, and open items. For example, an agent can remember how a customer wants a report formatted, then reuse that preference the next time the customer reaches out, regardless of which channel they use.

​
Conversation lifecycle

When a conversation ends (via timeout, end-call node, or opt-out), the agent stops responding. Restart by deleting the conversation or its messages in the messaging dashboard, then start a new thread. Per-number and per-request settings let you control:
  • Timeouts. When the user goes silent for a configured period, the agent can send a warning message and end the conversation.
  • Restart on reply. When restart_after_end_call is enabled, a new conversation starts automatically if the user replies after the previous one ended.
  • Webhooks. Each agent reply (and the conversation status) can be POSTed to a URL of your choice with the full message context.

​
Outcomes, citations, and summaries

Messaging conversations support the same post-conversation analytics as voice. When a conversation ends, Bland can:
  • Run outcomes to classify what happened.
  • Extract citations from the transcript.
  • Generate a summary using a custom prompt.
These can be configured per number on the messaging config, or passed per request on /sms/send. Webhook payloads include a channel field set to "sms", "rcs", "whatsapp", or "imessage" so you can route them to the right handler. When a conversation ends, the status webhook includes the full conversation context:
  • variables (pathway variables collected during the conversation)
  • metadata (custom metadata passed when the conversation was created)
  • pathway_tags (tags from pathway nodes visited during the conversation)
  • concatenated_transcript (the full conversation transcript as a single string)
  • summary (AI-generated summary, when a summary prompt is configured)
  • citations (extracted citation variables, when citation schemas are configured)
  • dispositions (outcome run results, when outcomes are configured)
Message webhooks also include variables, metadata, pathway_tags, and created_at for parity with voice call webhooks.

​
API

All three channels share the same endpoints. Pass channel in the request body, or let it be auto-derived from the number’s configuration.

Send a message

Send a message or start a conversation on any channel.

Create conversation

Create or resume a conversation with a specific pathway, node, and version.

List numbers

List your numbers and their channel configuration.

Conversations

List, retrieve, patch, and delete conversations.
For webhooks, delivery status, and number updates, see the Messaging section in the API Reference.
  • SMS (A2P registration and bring-your-own-Twilio)
  • RCS (verified business profile and rich messaging)
  • iMessage (branded iMessage agents with rich media)
  • Personas (one agent across voice and every messaging channel)
  • Memory (cross-channel contact recognition)
  • Outcomes (classify what happened in each conversation)
  • Post-Call Webhooks (webhook payloads for voice and messaging)
  • Billing (SMS pricing and usage)
Docs for agents: llms.txt