README.md

Build, run, and manage agent platforms.

Introduction

Agno is an SDK for building agent platforms.

Build agents using any framework. Run them as production services with sessions, memory, tracing, scheduling, and RBAC. Manage everything from a single control plane.

Here's what you can build:

• Coda → A code companion that lives in Slack and works alongside your team.
• Dash → A self-learning data agent that grounds answers in 6 layers of context.
• Scout → A context agent that navigates Slack and Google Drive to answer questions.
• Auto Improving Agent Platform → The leanest agent platform with a built-in auto-improvement loop.

Architecture

Agno has a 3-layer architecture. Everything except the control plane is free and open-source.

Quickstart

Run a coding agent as a service.

Built with the Agno SDK

Save this as workbench.py:

from agno.agent import Agent
from agno.db.sqlite import SqliteDb
from agno.os import AgentOS
from agno.tools.workspace import Workspace

workbench = Agent(
    name="Workbench",
    model="openai:gpt-5.4",
    tools=[Workspace(".",
        allowed=["read", "list", "search"],
        confirm=["write", "edit", "delete", "shell"],
    )],
    enable_agentic_memory=True,
    add_history_to_context=True,
    num_history_runs=3,
)

agent_os = AgentOS(
    agents=[workbench],
    tracing=True,
    db=SqliteDb(db_file="agno.db"),
)
app = agent_os.get_app()

Workspace(".") scopes the agent to the current directory. read, list, and search run freely. write, edit, delete, and shell require human approval.

from agno.agents.claude import ClaudeAgent
from agno.db.sqlite import SqliteDb
from agno.os import AgentOS

agent = ClaudeAgent(
    name="Claude Agent",
    model="claude-opus-4-7",
    allowed_tools=["Read", "Bash"],
    permission_mode="acceptEdits",
)

agent_os = AgentOS(agents=[agent], db=SqliteDb(db_file="agno.db"), tracing=True)
app = agent_os.get_app()

Run it

uv pip install -U 'agno[os]' openai

export OPENAI_API_KEY=sk-***

fastapi dev workbench.py

In 30 lines of code, you get:

• A FastAPI based agent server with 50+ endpoints
• Streaming responses, persistent sessions, per-user isolation
• Cron scheduling, human approval flows, and RBAC
• Native OpenTelemetry tracing

The API is available at http://localhost:8000. Checkout the OpenAPI spec at http://localhost:8000/docs.

Manage your platform using the AgentOS UI

The AgentOS UI provides a control plane that connects directly to your running AgentOS. Test agents, inspect runs, view traces, manage sessions, and monitor system health.

1. Open os.agno.com and sign in.
2. Click "Connect OS".
3. Select "Local".
4. Enter your endpoint URL (default: http://localhost:8000).
5. Name it "Local AgentOS" and click "Connect".

Open Chat, select your agent, and ask:

Tell me about the project

The agent reads your workspace and answers grounded in what it actually finds. Try a follow-up like "create a NOTES.md with three key takeaways." The run pauses for your approval before the file is written, since write is in the confirm list.

https://github.com/user-attachments/assets/adb38f55-1d9d-463e-8ca9-966bb6bdc37a

Get started

• Read the docs
• Build your first agent
• Build an auto-improving agent platform

Advantages of building an agent-platform with Agno

• Production API. 50+ endpoints with SSE and websockets to build a product on top of your agent platform.
• Storage. Store sessions, memory, knowledge, and traces in your own database. Postgres for sessions and memory. ClickHouse for OLAP data like traces.
• 100+ integrations. Pre-built toolkits for 100+ tools.
• Context Providers. Access live data from Slack, Drive, wikis, MCP, and custom sources.
• Human approval. Pause runs for user confirmation. Block tools that require admin approval.
• Observability. OpenTelemetry tracing, run history, and audit logs out of the box.
• Security. JWT-based RBAC and multi-user, multi-tenant isolation out of the box.
• Interfaces. Expose agents via Slack, Telegram, WhatsApp, Discord, AG-UI, A2A.
• Scheduling. Cron-based scheduling and background jobs with no external infrastructure.
• Deploy anywhere. Run on any cloud platform that runs containers. Docker, Railway, AWS, GCP.

Use Agno with your coding agent

Two options:

1. Add Agno docs as an indexed source. In Cursor: Settings → Indexing & Docs → Add https://docs.agno.com/llms-full.txt. Also works in VSCode, Windsurf, and similar tools.
2. Add Agno docs as an MCP server. Add docs.agno.com/mcp to your favourite coding agent.

Read the full guide here.

Community

• X / Twitter — follow for releases and demos
• Newsletter — monthly updates on what's shipping

Contributing

See the contributing guide.

Telemetry

Agno logs which model providers are used to prioritize updates. Disable with AGNO_TELEMETRY=false.

↑ Back to top