chester-ai 0.1.3

Python SDK for the Chester AI agent platform

chester-ai

Python SDK for the Chester AI agent platform. Full typed access to all 27 gRPC services.

Install

pip install chester-ai
# or
uv add chester-ai

Quick Start

from chester_ai import ChesterClient, proto

chester = ChesterClient(url="localhost:8990", api_key="cht_your_api_key_here")

# List all agents
response = chester.agents.ListAgents(proto.ListAgentsRequest())
for agent in response.agents:
    print(agent.name)

# Send a task to an agent
response = chester.queue.SendTask(proto.SendTaskRequest(
    agent="mybot",
    message="Summarize today's news",
))
print("Task queued:", response.task_id)

# Stream chat responses
for chunk in chester.agents.ChatWeb(proto.ChatStreamRequest(
    agent="mybot",
    user_message=proto.ChatUserMessage(content="Hello!"),
)):
    if chunk.text_delta.text:
        print(chunk.text_delta.text, end="")

# Subscribe to real-time events
for event in chester.events.SubscribeAll(proto.SubscribeAllRequest()):
    print(event.domain, event.event_type)

Async Usage

import asyncio
from chester_ai import AsyncChesterClient, proto

async def main():
    async with AsyncChesterClient(url="localhost:8990", api_key="cht_abc123") as chester:
        response = await chester.agents.ListAgents(proto.ListAgentsRequest())
        for agent in response.agents:
            print(agent.name)

asyncio.run(main())

Authentication

Chester uses Bearer tokens. Create one via the CLI:

chester auth create-token --name "my-app" --role operator

Services

All 27 Chester services are available as typed properties:

chester.agents, chester.queue, chester.decisions, chester.cron, chester.plans, chester.teams, chester.skills, chester.mcp, chester.daemon, chester.events, chester.context, chester.tools, chester.db, chester.temporal, chester.auth, chester.users, chester.goals, chester.capabilities, chester.templates, chester.organizations, chester.workflow_templates, chester.guardrails, chester.kanban, chester.observe, chester.datasets, chester.twins, chester.integrations

Requirements

• Python 3.10+
• Chester daemon running (default port 8990)