continuum-quickstart

Continuum Quickstart Skill

Use this skill when the user is starting from scratch. The goal is to get them to a running agent in a few minutes.

Path A — Library consumer

# 1. Create venv (Python 3.13 required) & install the package
python3.13 -m venv .venv && source .venv/bin/activate
pip install shyftlabs-continuum

# 2. Start infra (Redis :6380, Qdrant :6333) — `continuum up` ships with the
#    package, resolves the bundled stack, and writes a managed ./.env
continuum up

# 3. Add your provider key
echo "OPENAI_API_KEY=sk-..." >> .env

# 4. Run the smallest example (see snippet below)
python my_first_agent.py

Path B — Framework contributor (clone + editable install)

git clone https://github.com/bhavik-shyftlabs/continuum.git
cd continuum
cp .env.template .env                       # add OPENAI_API_KEY
continuum up                                # Redis + Qdrant (minimal); `continuum up full` adds Langfuse · Temporal · Milvus
python3.13 -m venv .venv && source .venv/bin/activate
pip install -e ".[dev,temporal,eval]"      # editable install with all extras

# Smoke-test against a runnable example app
python -m playground.sdk_feature_test

Smallest possible agent

import asyncio
from continuum.agent import BaseAgent, AgentRunner
from continuum.agent.config import AgentMemoryConfig
from continuum.core.container import Container, ContainerConfig

async def main():
    # No infra needed for this example
    container = Container(ContainerConfig(
        enable_memory=False, enable_session=False, enable_langfuse=False,
    ))
    agent = BaseAgent(
        name="hello",
        instructions="Reply in one short sentence.",
        model="gpt-4o-mini",
        memory_config=AgentMemoryConfig(search_memories=False, store_memories=False),
    )
    response = await AgentRunner(container=container).run(agent, "Say hi.")
    print(response.content)

asyncio.run(main())

Quick checks if something fails