OSS Tanbou
View on GitHubProject website

treat agent development and the production runtime as one platform stack

OSS scale score 315.9OSS health 93Update pace 124 commits/30d
About these scores

OSS scale score is an unbounded metric that log-compresses and weights Stars, Watchers, Forks, and Contributors. Discovery score is the current OSS scale score minus the score at discovery. Update pace is commits in the last 30 days, growth momentum is the OSS scale score difference within the recent observation window, and OSS health is a 0–100 rating based on available recency, Community Health, and release data.

Stars
42,406
Primary language
Python
License
Apache-2.0
Repository last updated
Sep 30, 2026
On this page

Overview

Agno provides a Python SDK for building agents and an AgentOS runtime for serving them through APIs and managing the platform through a UI. Sessions, memory, knowledge, MCP, human approval, scheduling, observability, and RBAC can be part of the production runtime.

Features and best fit

Based on official documentation; not hands-on tested · Content checked:

Key features

Build agents, teams, and workflows with a Python SDK

Models, tools, knowledge, and memory can be composed into individual agents, teams, and application workflows.

Sources: [2]

Use AgentOS for APIs, storage, MCP, and RBAC

The runtime provides more than 50 API endpoints, SSE and WebSockets, database storage, an MCP server, JWT-based RBAC, and multi-tenant isolation.

Sources: [2]

Best fit

Fits teams growing prototypes into self-managed agent platforms

It is useful when agent logic, deployment, storage, approvals, observability, and interfaces should share one runtime model.

Sources: [2]

Before adoption

Design least-privilege credentials around integrations

With integrations for GitHub, Slack, databases, MCP, and many other tools, agent tool credentials and human-approval boundaries should be deliberately minimized.

Sources: [2]

Disable run telemetry when required

The README says one telemetry event is sent per agent run while prompts, messages, and outputs are not transmitted. Set AGNO_TELEMETRY=false when no telemetry should be emitted.

Sources: [2]

Official sources

  1. [1]agno-agi/agno repository(2026-09-30)
  2. [2]Agno README(2026-09-30)
  3. [3]Agno Apache-2.0 license(2026-09-30)
  4. [4]Agno v3.0.11 release(2026-09-30)
  5. [5]Agno Python package metadata(2026-09-30)
Supplemental curator note

Agno is an Apache-2.0 agent-platform framework that goes beyond an agent SDK: AgentOS provides REST/SSE/WebSocket APIs, storage, MCP, scheduling, observability, and JWT-based RBAC. The README states that one telemetry event is sent per agent run, without prompts, messages, or outputs; set AGNO_TELEMETRY=false to disable it.

Try it in 3 steps

  1. 1

    Create a Python environment

    Create an isolated Python 3.9+ environment.

    python3 -m venv .venv && source .venv/bin/activate
  2. 2

    Install Agno 3.0.11

    Install the pinned Agno package and a model-provider client for the first agent.

    python -m pip install "agno==3.0.11" openai
  3. 3

    Run a minimal agent

    Replace the API key with your own and run a minimal agent with telemetry disabled.

    OPENAI_API_KEY=YOUR_KEY python -c "from agno.agent import Agent; from agno.models.openai import OpenAIChat; Agent(model=OpenAIChat(id="gpt-4o-mini"), telemetry=False).print_response("Say hello in one sentence")"
Check the official README

Growth

Growth trends · Last 30 days

42,406 Stars

Trend data is still being collected.

Development activity

Last 90 days · weekly

Commits (last 30 days)
124
Open PRs
902

Development activity is still being collected.

Built with

Categories and tags

Categories

Tags

GitHub data

GitHub dataView detailed GitHub data
View on GitHub

GitHub Topics

  • developer-tools
  • python
  • agents
  • ai
  • ai-agents
Stars
42,406
Forks
6,050
Watchers
238
Open issues
810
Contributors
440
Owner type
Organization
Primary language
Python
License
Apache-2.0
Repository last updated
Sep 30, 2026
Write a related article

Share a guide or use case for this OSS in Markdown. Articles are published after administrator approval.

Report incorrect information

Tell us if any listing information is incorrect or outdated.

After reading this page, do you know what to do next?