Agno with PraisonAI

Run PraisonAI agents YAML with the Agno framework via praisonai-frameworks

framework: agno connects PraisonAI’s YAML and CLI to the Agno multi-agent framework, routing tasks through Agno’s Team(mode=route) with a single install.

Quick Start

from praisonaiagents import Agent

agent = Agent(
    name="Helper",
    instructions="Answer briefly.",
    llm="openai/gpt-4o-mini",
)

agent.start("Reply with exactly OK.")

The same task expressed as a framework: agno YAML runs through Agno’s Team(mode=route) router — keep reading for YAML.

1. Install

   pip install "praisonai[agno]"

2. Set API Key

   export OPENAI_API_KEY=your-key

3. Create agents.yaml

   framework: agno
   topic: Quick task
   roles:
     helper:
       role: Helper
       goal: Answer briefly
       backstory: Helpful assistant
       tasks:
         answer:
           description: Reply with exactly OK.
           expected_output: OK

4. Run

   praisonai agents.yaml --framework agno

How Agno Works

Agno wraps each role into an agno.Agent, then assembles a Team(mode=route) that routes tasks to the appropriate specialist. For handoff topologies, the router agent delegates to sub-agents at runtime.

Capabilities

Sequential Context (Task Chaining)

Tasks reference outputs of earlier tasks with context: [task_name]:

framework: agno
topic: numbers
roles:
  writer:
    role: Writer
    goal: Write numbers only
    backstory: Concise writer
    tasks:
      draft:
        description: Reply with only the number 3.
        expected_output: "3"
      polish:
        description: Add 3 to the previous result. Reply with only the number.
        expected_output: "6"
        context:
          - draft

Multi-task configs without handoff.to run sequentially; context wires task outputs as inputs to subsequent tasks.

Handoffs

Use handoff.to with a single router task; specialists are routed via Agno Team(mode=route):

framework: agno
topic: language help
roles:
  triage:
    role: Triage Agent
    handoff:
      to:
        - English Agent
    tasks:
      route:
        description: Help with {topic}
  english:
    role: English Agent
    goal: Reply in English only
    backstory: English specialist.

Multi-task configs with handoff.to fall back to sequential execution.

Direct Adapter Use (Advanced)

Call the adapter without the CLI or YAML loader:

from praisonai_frameworks.agno.adapter import AgnoAdapter

config = {
    "framework": "agno",
    "topic": "Quick test",
    "roles": {
        "helper": {
            "role": "Assistant",
            "goal": "Answer briefly",
            "backstory": "Helpful assistant",
            "tasks": {
                "answer": {
                    "description": "Reply with exactly the word OK.",
                    "expected_output": "OK",
                }
            },
        }
    },
}
llm_config = [{"model": "openai/gpt-4o-mini", "api_key": ""}]   # use $OPENAI_API_KEY
result = AgnoAdapter().run(config, llm_config, "Quick test", tools_dict={})

Verify Installation

Check availability via praisonai doctor:

$ praisonai doctor
✓ Runtime 'agno' available
  name: Agno
  capabilities: agent_creation, tool_execution, sequential_execution, handoff, tool_loop

Or check from Python:

from praisonai._framework_availability import is_available

if is_available("agno"):
    print("Agno is installed and importable")

_agno_probe() checks three things in order:

1. importlib.metadata.distribution("agno") — the PyPI dist must be installed.
2. importlib.util.find_spec("agno") — the agno import namespace must be discoverable.
3. from agno.agent import Agent — the Agno Agent symbol must import without error.

True from is_available("agno") guarantees the adapter can run, not just that the package is on disk.

Pip Extras Reference

Troubleshooting

framework='agno' is not a valid choice — you are running a pre-PR-#2501 PraisonAI version. Upgrade: pip install -U praisonai.

Framework 'agno' was requested but is not installed — run pip install 'praisonai[agno]' (from your project rather than praisonai-frameworks[agno] alone, so praisonai-tools is also pulled in).

ValueError: framework='agno' in workflow YAML is not supported for workflow execution — switch the file to the roles: agents.yaml shape used in the Quick Start above.

agno imports successfully but is_available("agno") returns False — the probe also requires agno to be visible to importlib.metadata.distribution(...). If the package was installed in editable or namespace-only mode without a real dist-info, the probe refuses. Install the published agno PyPI package.

Best Practices

Related