Skip to main content

Debug UI for Microsoft Agent Framework with OpenAI-compatible API server.

Project description

DevUI - A Sample App for Running Agents and Workflows

A lightweight, standalone sample app interface for running entities (agents/workflows) in the Microsoft Agent Framework supporting directory-based discovery, in-memory entity registration, and sample entity gallery.

[!IMPORTANT] DevUI is a sample app to help you get started with the Agent Framework. It is not intended for production use. For production, or for features beyond what is provided in this sample app, it is recommended that you build your own custom interface and API server using the Agent Framework SDK.

DevUI Screenshot

Quick Start

# Install
pip install agent-framework-devui

You can also launch it programmatically

from agent_framework import ChatAgent
from agent_framework.openai import OpenAIChatClient
from agent_framework.devui import serve

def get_weather(location: str) -> str:
    """Get weather for a location."""
    return f"Weather in {location}: 72°F and sunny"

# Create your agent
agent = ChatAgent(
    name="WeatherAgent",
    chat_client=OpenAIChatClient(),
    tools=[get_weather]
)

# Launch debug UI - that's it!
serve(entities=[agent], auto_open=True)
# → Opens browser to http://localhost:8080

In addition, if you have agents/workflows defined in a specific directory structure (see below), you can launch DevUI from the cli to discover and run them.

# Launch web UI + API server
devui ./agents --port 8080
# → Web UI: http://localhost:8080
# → API: http://localhost:8080/v1/*

When DevUI starts with no discovered entities, it displays a sample entity gallery with curated examples from the Agent Framework repository to help you get started quickly.

Directory Structure

For your agents to be discovered by the DevUI, they must be organized in a directory structure like below. Each agent/workflow must have an __init__.py that exports the required variable (agent or workflow).

Note: .env files are optional but will be automatically loaded if present in the agent/workflow directory or parent entities directory. Use them to store API keys, configuration variables, and other environment-specific settings.

agents/
├── weather_agent/
│   ├── __init__.py      # Must export: agent = ChatAgent(...)
│   ├── agent.py
│   └── .env             # Optional: API keys, config vars
├── my_workflow/
│   ├── __init__.py      # Must export: workflow = WorkflowBuilder()...
│   ├── workflow.py
│   └── .env             # Optional: environment variables
└── .env                 # Optional: shared environment variables

Viewing Telemetry (Otel Traces) in DevUI

Agent Framework emits OpenTelemetry (Otel) traces for various operations. You can view these traces in DevUI by enabling tracing when starting the server.

devui ./agents --tracing framework

OpenAI-Compatible API

For convenience, you can interact with the agents/workflows using the standard OpenAI API format. Just specify the entity_id in the extra_body field. This can be an agent_id or workflow_id.

# Standard OpenAI format
curl -X POST http://localhost:8080/v1/responses \
  -H "Content-Type: application/json" \
  -d @- << 'EOF'
{
  "model": "agent-framework",
  "input": "Hello world",
  "extra_body": {"entity_id": "weather_agent"}
}

CLI Options

devui [directory] [options]

Options:
  --port, -p      Port (default: 8080)
  --host          Host (default: 127.0.0.1)
  --headless      API only, no UI
  --config        YAML config file
  --tracing       none|framework|workflow|all
  --reload        Enable auto-reload

Key Endpoints

  • GET /v1/entities - List discovered agents/workflows
  • GET /v1/entities/{entity_id}/info - Get detailed entity information
  • POST /v1/entities/add - Add entity from URL (for gallery samples)
  • DELETE /v1/entities/{entity_id} - Remove remote entity
  • POST /v1/responses - Execute agent/workflow (streaming or sync)
  • GET /health - Health check
  • POST /v1/threads - Create thread for agent (optional)
  • GET /v1/threads?agent_id={id} - List threads for agent
  • GET /v1/threads/{thread_id} - Get thread info
  • DELETE /v1/threads/{thread_id} - Delete thread
  • GET /v1/threads/{thread_id}/messages - Get thread messages

Implementation

  • Discovery: agent_framework_devui/_discovery.py
  • Execution: agent_framework_devui/_executor.py
  • Message Mapping: agent_framework_devui/_mapper.py
  • Session Management: agent_framework_devui/_session.py
  • API Server: agent_framework_devui/_server.py
  • CLI: agent_framework_devui/_cli.py

Examples

See samples/ for working agent and workflow implementations.

License

MIT

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

agent_framework_devui-1.0.0b251001.tar.gz (615.8 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

agent_framework_devui-1.0.0b251001-py3-none-any.whl (256.7 kB view details)

Uploaded Python 3

File details

Details for the file agent_framework_devui-1.0.0b251001.tar.gz.

File metadata

File hashes

Hashes for agent_framework_devui-1.0.0b251001.tar.gz
Algorithm Hash digest
SHA256 891368951673b5fe2f7f25174eeab032d9faed37cf06d82a5f91ee5cdd8b1c15
MD5 c13fd534b10cea0feb1bec0b9b6019ff
BLAKE2b-256 51ceee2d16360cc6de9214bfb11921b52a4598fd3f3789521204d499a8907a4e

See more details on using hashes here.

File details

Details for the file agent_framework_devui-1.0.0b251001-py3-none-any.whl.

File metadata

File hashes

Hashes for agent_framework_devui-1.0.0b251001-py3-none-any.whl
Algorithm Hash digest
SHA256 57b39fe39c2752b3df18941c581d12c728b5e5ad5814bad66d186db932449edb
MD5 167bfd042e4216b1d1ca512330bf4f93
BLAKE2b-256 7b43a124de5659272e967e75cdcdea407facf3e6b94becf0b1218297ed59d502

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page