a2a-lite 0.1.0

Simplified wrapper for Google's A2A Protocol SDK

A2A Lite 🚀

Build A2A agents in 8 lines. Add enterprise features when you need them.

Wraps the official A2A SDKs (Python, TypeScript, Java) with a simple, intuitive API.

Python	TypeScript	Java
from a2a_lite import Agent agent = Agent( name="Bot", description="My bot" ) @agent.skill("greet") async def greet(name: str): return f"Hello, {name}!" agent.run()	import { Agent } from 'a2a-lite'; const agent = new Agent({ name: 'Bot', description: 'My bot' }); agent.skill('greet', async ({ name }) => `Hello, ${name}!` ); agent.run();	var agent = Agent.builder() .name("Bot") .description("My bot") .build(); agent.skill("greet", params -> "Hello, " + params.get("name") + "!" ); agent.run();

That's it. You have a running A2A agent.

---

Installation

Python

pip install a2a-lite

TypeScript

npm install a2a-lite
# or
yarn add a2a-lite

Java (Gradle)

dependencies {
    implementation 'com.a2alite:a2a-lite:0.1.0'
    implementation 'io.javalin:javalin:5.6.3'  // For standalone mode
}

Java (Maven)

<dependency>
    <groupId>com.a2alite</groupId>
    <artifactId>a2a-lite</artifactId>
    <version>0.1.0</version>
</dependency>

---

The Philosophy

Need	A2A Lite
Hello World	8 lines
Testing	3 lines
Pydantic models	Just works
Auth	Add when needed
Files	Add when needed
Human-in-the-loop	Add when needed
Task tracking	Add when needed

Everything is optional except the basics.

---

Progressive Complexity

Level 1: Just Works (No Setup)

from a2a_lite import Agent

agent = Agent(name="Bot", description="A bot")

@agent.skill("greet")
async def greet(name: str) -> str:
    return f"Hello, {name}!"

agent.run()

Level 2: Use Pydantic (Just Works)

from pydantic import BaseModel

class User(BaseModel):
    name: str
    email: str

@agent.skill("create_user")
async def create_user(user: User) -> dict:
    # 'user' is already a User instance - auto-converted!
    return {"id": 1, "name": user.name}

Level 3: Stream Responses (Just Yield)

@agent.skill("chat", streaming=True)
async def chat(message: str):
    for word in message.split():
        yield word + " "

Level 4: Add Middleware (When Needed)

@agent.middleware
async def log_requests(ctx, next):
    print(f"Calling: {ctx.skill}")
    return await next()

Level 5: Human-in-the-Loop (When Needed)

from a2a_lite import InteractionContext

@agent.skill("wizard")
async def wizard(ctx: InteractionContext) -> dict:
    name = await ctx.ask("What's your name?")
    role = await ctx.ask("Role?", options=["Dev", "Manager"])

    if await ctx.confirm(f"Create {name} as {role}?"):
        return {"created": name}
    return {"cancelled": True}

Level 6: Handle Files (When Needed)

from a2a_lite import FilePart

@agent.skill("summarize")
async def summarize(doc: FilePart) -> str:
    content = await doc.read_text()
    return f"Summary: {content[:100]}..."

Level 7: Track Task Progress (When Needed)

from a2a_lite import TaskContext

agent = Agent(name="Bot", task_store="memory")

@agent.skill("process")
async def process(data: str, task: TaskContext) -> str:
    await task.update("working", "Starting...", progress=0.0)

    for i in range(10):
        await task.update("working", f"Step {i}/10", progress=i/10)

    return "Done!"

Level 8: Add Authentication (When Needed)

from a2a_lite import Agent, APIKeyAuth

agent = Agent(
    name="SecureBot",
    auth=APIKeyAuth(keys=["secret-key"]),
)

---

Testing (3 Lines)

Python	TypeScript	Java
from a2a_lite import TestClient client = TestClient(agent) assert client.call("greet", name="World") == "Hello, World!"	import { TestClient } from 'a2a-lite'; const client = new TestClient(agent); expect(await client.call('greet', { name: 'World' })).toBe('Hello, World!');	var client = new TestClient(agent); assertThat(client.call("greet", Map.of("name", "World"))) .isEqualTo("Hello, World!");

---

CLI

a2a-lite init my-agent          # Create new project
a2a-lite inspect http://...     # View agent capabilities
a2a-lite test http://... skill  # Test a skill
a2a-lite discover               # Find local agents

---

Full Feature List

Feature	Import	Complexity
Basic skills	Agent	Just works
Pydantic models	BaseModel	Just works
Streaming	streaming=True	Just works
Testing	TestClient	Just works
Middleware	@agent.middleware	Add when needed
Webhooks	@agent.on_complete	Add when needed
Human-in-the-loop	InteractionContext	Add when needed
Conversation memory	ConversationMemory	Add when needed
File handling	FilePart	Add when needed
Structured data	DataPart	Add when needed
Rich outputs	Artifact	Add when needed
Task tracking	TaskContext	Add when needed
API Key auth	APIKeyAuth	Add when needed
Bearer/JWT auth	BearerAuth	Add when needed
Multiple auth	CompositeAuth	Add when needed

---

Examples

Example	What it shows
01_hello_world.py	Simplest agent (8 lines)
02_calculator.py	Multiple skills
06_pydantic_models.py	Auto Pydantic conversion
08_streaming.py	Streaming responses
09_testing.py	Testing your agents
11_human_in_the_loop.py	Ask user questions
12_file_handling.py	Handle files
13_task_tracking.py	Progress updates
14_with_auth.py	Authentication

---

For AI Coding Assistants

See AGENT.md for a quick reference guide optimized for AI coding assistants implementing A2A agents.

---

100% A2A Protocol Compatible

A2A Lite wraps the official A2A SDKs. Every feature maps to real A2A protocol concepts:

A2A Lite	A2A Protocol
@agent.skill() / agent.skill()	Agent Skills
streaming=True / SkillConfig.withStreaming()	SSE Streaming
InteractionContext.ask()	input-required state
TaskContext.update()	Task lifecycle states
FilePart	A2A File parts
APIKeyAuth / BearerAuth	Security schemes

---

Language-Specific Documentation

Language	Package	Documentation
Python	a2a-lite	packages/python/README.md
TypeScript	a2a-lite	packages/typescript/README.md
Java	com.a2alite:a2a-lite	packages/java/README.md

---

License

Apache 2.0