> ## Documentation Index
> Fetch the complete documentation index at: https://docs.agno.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Studio Tools Agent

> Agent with StudioTools -- let the agent compose new agents, teams, and workflows.

```python studio_tools_agent.py theme={null}
"""Agent with StudioTools -- let the agent compose new agents, teams, and workflows.

The StudioTools uses the AgentOS Registry (tools, models, dbs) and the core
component APIs (Agent, Team, Workflow, Step) to dynamically build new
components described by the user in natural language.

Ask the studio agent things like:
    - "What models and tools do we have available?"
    - "Create an agent named 'news' using claude-sonnet-4-6 with DuckDuckGoTools
       that summarizes news headlines in 2-3 sentences."
    - "Create a team called 'research' with the news agent and the Greeter agent."
    - "Create a workflow called 'daily-briefing' that runs the news agent then the
       Reporter agent."
    - "Run the news agent with message 'Top AI story today?'"

Usage:
    # Start the AgentOS server
    .venvs/demo/bin/python cookbook/05_agent_os/studio_tool/studio_tools_agent.py
"""

from pathlib import Path

from agno.agent import Agent
from agno.db.sqlite import SqliteDb
from agno.models.anthropic import Claude
from agno.models.openai import OpenAIResponses
from agno.os import AgentOS
from agno.registry import Registry
from agno.tools.calculator import CalculatorTools
from agno.tools.duckduckgo import DuckDuckGoTools
from agno.tools.hackernews import HackerNewsTools
from agno.tools.studio import StudioTools

DB_DIR = Path(__file__).parent / "tmp"
DB_DIR.mkdir(exist_ok=True)

db = SqliteDb(id="studio-demo2-db", db_file=str(DB_DIR / "studio_demo2.db"))

registry = Registry(
    name="Studio Registry",
    tools=[DuckDuckGoTools(), HackerNewsTools(), CalculatorTools()],
    models=[
        OpenAIResponses(id="gpt-5.5"),
        Claude(id="claude-sonnet-4-6"),
    ],
    dbs=[db],
)

greeter = Agent(
    id="greeter",
    name="Greeter",
    model=OpenAIResponses(id="gpt-5.5"),
    instructions=["You are a friendly greeter."],
    db=db,
)

reporter = Agent(
    id="reporter",
    name="Reporter",
    model=OpenAIResponses(id="gpt-5.5"),
    instructions=["You summarize news headlines in 2-3 sentences."],
    db=db,
)


studio_agent = Agent(
    id="studio-agent",
    name="Studio Agent",
    model=OpenAIResponses(id="gpt-5.5"),
    tools=[
        StudioTools(
            registry=registry,
            db=db,
            agents_list=[greeter, reporter],
            default_model_id="gpt-5.5",
            versions=True,
        ),
    ],
    instructions=[
        "You are an AgentOS studio. You help users compose new agents, teams, and workflows.",
        "Always start by listing the available models, tools, and existing agents so the user knows what primitives are on hand.",
        "Before calling create_agent, restate the exact tool names you plan to pass and confirm they match what the user requested. If the user named a tool (e.g. 'calculator'), you MUST include that exact name in tool_names.",
        "After creating or editing a component, respond with component_type, component_id, name, db_version or draft_version, and the next action.",
        "Include the Studio route for the component: agents use /studio/agents/edit?agent_id=<id>, teams use /studio/teams/edit?team_id=<id>, workflows use /studio/workflows/edit?workflow_id=<id>, and registry primitives live at /studio/registry.",
        "Do not include a Studio component link when the tool returned an error.",
    ],
    db=db,
    markdown=True,
)


agent_os = AgentOS(
    id="studio-agent-os",
    agents=[greeter, reporter, studio_agent],
    registry=registry,
    db=db,
)
app = agent_os.get_app()


if __name__ == "__main__":
    agent_os.serve(app="studio_tools_agent:app", port=7777, reload=True)
```

## Run the Example

<Steps>
  <Snippet file="create-venv-step.mdx" />

  <Step title="Install dependencies">
    ```bash theme={null}
    uv pip install -U "agno[os]" anthropic ddgs fastmcp openai starlette
    ```
  </Step>

  <Step title="Export your API keys">
    <CodeGroup>
      ```bash Mac/Linux theme={null}
      export ANTHROPIC_API_KEY="your_anthropic_api_key_here"
      export JWT_VERIFICATION_KEY="your_jwt_verification_key_here"
      export OPENAI_API_KEY="your_openai_api_key_here"
      ```

      ```bash Windows theme={null}
      $Env:ANTHROPIC_API_KEY="your_anthropic_api_key_here"
      $Env:JWT_VERIFICATION_KEY="your_jwt_verification_key_here"
      $Env:OPENAI_API_KEY="your_openai_api_key_here"
      ```
    </CodeGroup>
  </Step>

  <Step title="Run the example">
    Save the code above as `studio_tools_agent.py`, then run:

    ```bash theme={null}
    python studio_tools_agent.py
    ```
  </Step>
</Steps>

Full source: [cookbook/05\_agent\_os/studio\_tool/studio\_tools\_agent.py](https://github.com/agno-agi/agno/blob/main/cookbook/05_agent_os/studio_tool/studio_tools_agent.py)