> ## 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.

# Metadata Resolution

> Demonstrates the three-layer metadata resolution for Agents: agent.metadata < session.metadata < call-site metadata.

```python metadata_resolution.py theme={null}
"""
Metadata Resolution
=============================

Demonstrates the three-layer metadata resolution for Agents:
  agent.metadata < session.metadata < call-site metadata

Session-stored metadata overrides agent defaults, and call-site
metadata overrides both. This works for both sync and async runs.
"""

import asyncio
import time

from agno.agent import Agent
from agno.db.in_memory import InMemoryDb
from agno.models.openai import OpenAIResponses
from agno.session.agent import AgentSession

# ---------------------------------------------------------------------------
# Setup
# ---------------------------------------------------------------------------
db = InMemoryDb()

# Pre-seed a session with metadata (simulates a returning user)
session = AgentSession(
    session_id="demo-session",
    agent_id="metadata-demo-agent",
    metadata={
        "user_tier": "premium",  # Will override agent's "free"
        "session_pref": "dark_mode",  # Session-only key
    },
    created_at=int(time.time()),
)
db.upsert_session(session)

# ---------------------------------------------------------------------------
# Create Agent
# ---------------------------------------------------------------------------
agent = Agent(
    id="metadata-demo-agent",
    model=OpenAIResponses(id="gpt-5-mini"),
    db=db,
    metadata={
        "user_tier": "free",  # Will be overridden by session
        "agent_env": "production",  # Agent-only key
    },
)


# ---------------------------------------------------------------------------
# Run Agent
# ---------------------------------------------------------------------------
async def run_async_demo() -> None:
    print("\n=== Async run with all three layers ===")
    result = await agent.arun(
        input="Say 'async test'",
        session_id="demo-session",
        metadata={"user_tier": "vip"},
    )
    print(f"Metadata: {result.metadata}")
    print(f"  user_tier = '{result.metadata['user_tier']}' (call-site wins)")
    print(f"  session_pref = '{result.metadata['session_pref']}' (from session)")
    print(f"  agent_env = '{result.metadata['agent_env']}' (from agent)")


if __name__ == "__main__":
    # Session metadata overrides agent metadata
    print("=== Session overrides agent (sync) ===")
    result1 = agent.run(input="Say 'test1'", session_id="demo-session")
    print(f"Metadata: {result1.metadata}")
    print(f"  user_tier = '{result1.metadata['user_tier']}' (session wins over agent)")

    # Call-site metadata overrides both
    print("\n=== Call-site overrides all (sync) ===")
    result2 = agent.run(
        input="Say 'test2'",
        session_id="demo-session",
        metadata={"user_tier": "enterprise", "request_id": "req-123"},
    )
    print(f"Metadata: {result2.metadata}")
    print(f"  user_tier = '{result2.metadata['user_tier']}' (call-site wins)")

    # Async run
    asyncio.run(run_async_demo())

    # New session uses agent defaults
    print("\n=== New session uses agent defaults ===")
    result4 = agent.run(input="Say 'test4'", session_id="new-session")
    print(f"Metadata: {result4.metadata}")
    print(f"  user_tier = '{result4.metadata['user_tier']}' (agent default)")

    print("\n" + "=" * 50)
    print("Resolution order: agent < session < call-site")
    print("=" * 50)
```

## Run the Example

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

  <Step title="Install dependencies">
    ```bash theme={null}
    uv pip install -U agno openai
    ```
  </Step>

  <Step title="Export your OpenAI API key">
    <CodeGroup>
      ```bash Mac/Linux theme={null}
      export OPENAI_API_KEY="your_openai_api_key_here"
      ```

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

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

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

Full source: [cookbook/02\_agents/05\_state\_and\_session/metadata\_resolution.py](https://github.com/agno-agi/agno/blob/v3.0.4/cookbook/02_agents/05_state_and_session/metadata_resolution.py)