pattern-agentic-messaging 0.8.4

SLIM-powered messaging

Pattern Agentic Messaging

An async SLIM wrapper with a FastAPI-like interface

Installation

pip install pattern_agentic_messaging

Usage

Server

Route messages to decorated methods based on a discriminator field like type:

from pattern_agentic_messaging import PASlimApp, PASlimConfig
from .models import QuestionRequest, StatusRequest, AnswerResponse

config = PASlimConfig(
    local_name="org/ns/server/instance1",
    endpoint="https://slim.example.com",
    auth_secret="shared-secret",
    message_discriminator="type"
)

app = PASlimApp(config)
agent = None

@app.on_session_connect
async def on_connect(session):
    agent = await create_agent(...)
    session.context = {
        "agent": agent
    }
    

@app.on_message
async def handle_prompt(session, msg: QuestionRequest):
    agent = session.context.get("agent")
    response = await agent.ask(msg["question"])
    await session.send(AnswerResponse(answer=response))


@app.on_message
async def handle_status(session, msg: StatusRequest):
    await session.send({"type": "status", "value": "ready"})

@app.on_message
async def handle_other(session, msg):
    await session.send({"error": f"Unknown message type: {msg.get('type')}"})

app.run()

Use PASlimConfigGroup to create a group channel.

The models are pydantic models, which must have a literal field corresponding to the discriminator:

from pydantic import BaseModel

class QuestionRequest(BaseModel):
    type: Literal["question"] = "question"
    prompt: str


class StatusRequest(BaseModel):
    type: Literal["status"] = "status"


class AnswerResponse(BaseModel):
    type: Literal["answer"] = "answer"
    answer: str

Client

Connect to a specific peer:

from pattern_agentic_messaging import PASlimApp, PASlimConfig

config = PASlimConfig(
    local_name="org/ns/client/instance1",
    endpoint="https://slim.example.com",
    auth_secret="shared-secret"
)

async with PASlimApp(config) as app:
    async with await app.connect("org/ns/server/instance1") as session:
        await session.send({"type": "prompt", "prompt": "Hello world!"})
        async for msg in session:
            print(f"RECEIVED: {msg}")

Alternatively to join a group channel:

from pattern_agentic_messaging import PASlimApp, PASlimConfig

config = PASlimConfig(
    local_name="org/ns/participant/p1",
    endpoint="https://slim.example.com",
    auth_secret="shared-secret"
)

async with PASlimApp(config) as app:
    async with await app.join_channel() as session:
        async for msg in session:
            print(f"Channel message: {msg}")
            await session.send({"type": "response", "msg": "received"})

Low-level usage

The API behind the decorator pattern can be used directly:

from pattern_agentic_messaging import PASlimApp, PASlimConfig

config = PASlimConfig(
    local_name="org/ns/server/instance1",
    endpoint="https://slim.example.com",
    auth_secret="shared-secret"
)

async with PASlimApp(config) as app:
    async for session, msg in app:
        if not isinstance(msg, dict) or "prompt" not in msg:
            await session.send({"error": "Invalid format"})
            continue

        result = await process(msg["prompt"])
        await session.send({"result": result})