Skip to main content
The Unified Chat Interface provides a web-based UI for interacting with your agents. Test prompts, debug behavior, and share access with team members - all without writing code.

​
Overview

Real-time Chat

Interactive conversations with streaming responses

Tool Visibility

See tool calls and results as they happen

Session History

Browse and resume previous conversations

Share Access

Generate shareable links for stakeholders

​
Accessing the Chat Interface

​
Via Workbench

  1. Open your agent in the Workbench
  2. Click Chat in the top navigation
  3. Start chatting with your agent

​
Direct URL

Access any agent’s chat interface directly: 
https://app.xpander.ai/chat/{agent-id}

​
Features

​
Streaming Responses

Watch responses generate in real-time: 
User: Summarize today's GitHub issues

Agent: Let me check your GitHub issues...
       [Calling: github-issues-list]
       [Result: 12 issues found]

       Here's a summary of today's issues:

       **Critical (2)**
       - #423: Production database connection timeout
       - #421: Authentication failing for OAuth users
       ...

​
Tool Call Visibility

See exactly what tools are being invoked: 
┌─────────────────────────────────────┐
│ Tool: github-issues-list            │
│ Status: ✓ Completed (234ms)         │
├─────────────────────────────────────┤
│ Input:                              │
│   repository: "my-org/my-repo"      │
│   state: "open"                     │
│   labels: ["bug", "critical"]       │
├─────────────────────────────────────┤
│ Output:                             │
│   12 issues returned                │
│   [Expand to see full response]     │
└─────────────────────────────────────┘

​
File Attachments

Upload files directly in chat:
  • Documents (PDF, DOCX, TXT)
  • Images (PNG, JPG, GIF)
  • Data files (CSV, JSON)
User: [Attached: quarterly-report.pdf]
      Summarize the key findings from this report

Agent: I'll analyze the quarterly report...

​
Conversation History

Browse and resume previous sessions: 
Recent Conversations
─────────────────────────────────────
Today
  • "Debug authentication issue" - 2 hours ago
  • "Review PR #156" - 4 hours ago

Yesterday
  • "Generate weekly report" - 1 day ago
  • "Analyze error logs" - 1 day ago

This Week
  • "Plan Q2 roadmap" - 3 days ago

​
Session Management

​
Create New Session

from xpander_sdk import Chat

chat = Chat(agent_id="agent-123")
session = chat.create_session(
    title="Debug Session",
    metadata={"ticket": "JIRA-456"}
)

print(f"Session URL: {session.url}")

​
Resume Session

# Resume existing conversation
session = chat.get_session("session-789")
response = session.send("Continue from where we left off")

​
Export Conversation

# Export as markdown
markdown = session.export(format="markdown")

# Export as JSON
data = session.export(format="json")

​
Sharing & Access Control

Create shareable links with configurable permissions: 
from xpander_sdk import Chat

chat = Chat(agent_id="agent-123")
share_link = chat.create_share_link(
    session_id="session-789",
    permissions="read_only",  # or "interactive"
    expires_in="7d"
)

print(f"Share URL: {share_link.url}")

​
Permission Levels

PermissionDescription
read_onlyView conversation history only
interactiveCan send new messages
fullFull access including settings

​
Embed in Your App

Embed the chat interface in your application: 
<iframe
  src="https://app.xpander.ai/embed/chat/{agent-id}"
  width="400"
  height="600"
  frameborder="0"
></iframe>

​
Customization

​
Theme

chat = Chat(
    agent_id="agent-123",
    theme={
        "primary_color": "#4F46E5",
        "background": "#F9FAFB",
        "font_family": "Inter"
    }
)

​
Branding

Add custom branding in Workbench:
  1. Navigate to Agent Settings > Chat Interface
  2. Upload logo and configure colors
  3. Set custom welcome message

​
Welcome Message

chat = Chat(
    agent_id="agent-123",
    welcome_message="Hi! I'm your DevOps assistant. How can I help you today?",
    suggested_prompts=[
        "Check system status",
        "Review recent deployments",
        "Analyze error logs"
    ]
)

​
Keyboard Shortcuts

ShortcutAction
EnterSend message
Shift+EnterNew line
Ctrl+KNew conversation
Ctrl+/Show shortcuts
EscCancel current request

​
API Integration

Build custom chat experiences: 
from xpander_sdk import Backend
from agno.agent import Agent

backend = Backend()
agent = Agent(**backend.get_args())

# Streaming chat
for chunk in agent.run_stream("Hello, how can you help?"):
    print(chunk.content, end="", flush=True)

Web UI Task Source

Configure web interface

Slack Integration

Chat via Slack