# Agno (slim)

> Open-source Python framework from Agno Inc. for building agents, teams and workflows, with the AgentOS runtime that serves them over a REST API and an MCP server. Formerly Phidata.

- Full: https://www.anchorterminal.com/tools/agno.md (~6,000 tokens) · this version ~1,430 tokens · JSON https://www.anchorterminal.com/tools/agno.json · canonical https://www.anchorterminal.com/tools/agno
- Index: https://www.anchorterminal.com/llms.txt · API: https://www.anchorterminal.com/api/v1/index.json · Updated: 2026-10-08

**BB · 72.8/100 · rank #85 of 722 · #7 in Agent frameworks & SDKs · agent-ready · confidence medium**


Assessment: An Apache 2.0 Python agent framework at 3.1.1 with an MCP client, tool confirmation and admin approvals, a durable job queue and 25 stable releases in 90 days. Usage telemetry is on by default with unhashed identifiers, and three high or critical advisories were published against it in the last 12 months, all fixed.

## Facts

- Kind: Agent framework · vendor: Agno Inc. · category: Agent frameworks & SDKs · legal entity: Agno Inc. · provenance 93/100
- Packages: pypi `agno`
- Auth: API key · pricing: Free · x402: no · licence: Apache-2.0
- Languages: Python 3.9 or later
- Version: 3.1.1 on 2 October 2026. 3.0.0 on 24 August 2026
- Models: Provider modules for OpenAI, Anthropic, Google and others under agno.models (about 60 entries in the source directory)
- MCP client: MCPTools in the agno[mcp] extra. Streamable HTTP and stdio, with SSE marked deprecated. Tool filters, name prefix and dynamic headers
- MCP server: AgentOS with `mcp=True` serves /mcp over Streamable HTTP beside the REST API
- Multi-agent: Team, with a leader that delegates to member agents, and Workflow for sequential, parallel, conditional and looping steps
- Durable state: Sessions and runs stored in your database. AgentOS background runs survive restarts with QueueConfig(durable=True), off by default
- Human approval: requires_confirmation, user input, external execution and admin approvals with stored records, on agents, teams and workflow steps
- Guardrails: PII patterns, literal prompt-injection phrases and OpenAI moderation, as pre-hooks on the run input
- Tracing: OpenTelemetry, stored in your database or exported
- Telemetry: On by default. One event per run to os-api.agno.com with model, feature flags and unhashed identifiers. Off with AGNO_TELEMETRY=false or telemetry=False
- Hosted control plane: os.agno.com. Free for a local AgentOS, Pro $150 a month for one live connection and three seats
- Releases in 90 days: 25 stable and 5 pre-releases on PyPI
- Prices: Control plane Pro plan, one live AgentOS connection and three seats $150 per month (plan)

- Scores: Reliability 83, Performance pending, Schema & documentation 90, Agent ergonomics 76, Security & auth 64, Payments & pricing 50, Task success pending, Maintenance & community 86, Transparency & trust 83 · negative events -3 · total over the 7 assessed categories
- Why: Reliability, Official PyPI package with Python 3.9 or later stated (20). · Schema & documentation, A class reference under docs.agno.com/reference and a REST reference for AgentOS, which also serves /openapi.json (25). · Agent ergonomics, An agent with an MCP server is about ten lines through MCPTools, with include_tools, exclude_tools and tool_name_prefix (22). · Security & auth, Telemetry is on by default. · Payments & pricing, No payment protocol (0). · Maintenance & community, 3.1.1 on 2 October 2026 (30). · Transparency & trust, Apache 2.0 (30).
- Sources: 24, open questions: 7, both in the full twin
- Capabilities: agent.framework, agent.multi-agent, agent.durable, agent.mcp-client
- JSON: https://www.anchorterminal.com/api/v1/tools/agno.json
- Verify (for the vendor): the badge `https://www.anchorterminal.com/badges/agno.svg` or a link to https://www.anchorterminal.com/tools/agno from a page on agno.com or one of its subdomains, or the README of github.com/agno-agi/agno, then `POST https://www.anchorterminal.com/api/v1/verify` `{"slug", "url"}` or `verify_listing` at /mcp; re-checked weekly, no effect on the grade. Snippets in the full twin.

## Before you call it

1. Set AGNO_TELEMETRY=false before the first run, and pass telemetry=False to AgentOS and to each eval, which ignore the variable
2. Configure JWT_VERIFICATION_KEY with AgentOS authorisation switched on, or OS_SECURITY_KEY, before exposing AgentOS. With neither set, REST and MCP routes take no credential
3. Mark tools that write with requires_confirmation, and resume with continue_run once every requirement is resolved
4. Set tool_call_limit on agents. The default is None
5. Install extras for what you use, such as agno[mcp] for MCPTools and agno[os,mcp] for AgentOS with its MCP server

## Connect

```bash
pip install -U agno   # MCP client: pip install -U "agno[mcp]"
```

## Similar tools

| Tool | Grade | Score | Shared capabilities | Slim |
| --- | --- | --- | --- | --- |
| OpenAI Agents SDK | AA | 86.2 | agent.framework, agent.multi-agent, agent.durable, agent.mcp-client | https://www.anchorterminal.com/tools/openai-agents-sdk.min.md |
| Pydantic AI | A | 83.7 | agent.framework, agent.multi-agent, agent.durable, agent.mcp-client | https://www.anchorterminal.com/tools/pydantic-ai.min.md |
| Microsoft Agent Framework | A | 82.2 | agent.framework, agent.multi-agent, agent.durable, agent.mcp-client | https://www.anchorterminal.com/tools/microsoft-agent-framework.min.md |
| Docker Agent | BB | 76.5 | agent.framework, agent.multi-agent, agent.mcp-client, agent.durable | https://www.anchorterminal.com/tools/docker-agent.min.md |
| Agent Development Kit (ADK) | BB | 74.7 | agent.framework, agent.multi-agent, agent.durable, agent.mcp-client | https://www.anchorterminal.com/tools/google-adk.min.md |


## Panel reviews (0, desk reviews from public material, no calls made)