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

# Context Assembly

> Server-side context assembly combining short-term messages and long-term memories

The `context()` method assembles LLM-ready context by combining recent conversation messages (short-term) with semantically relevant memories (long-term) — all server-side. Use it instead of manually calling `search()` and formatting results.

## Basic Usage

<Tabs>
  <Tab title="Hosted (lyzr-adk)">
    ```python theme={null}
    from lyzr import Cognis, CognisMessage

    cog = Cognis()

    context = cog.context(
        current_messages=[
            CognisMessage(role="user", content="What should I do this weekend?"),
        ],
        owner_id="user_alice",
        session_id="sess_001",
    )
    # Use in your LLM prompt
    print(context)
    ```
  </Tab>

  <Tab title="Open Source (lyzr-cognis)">
    ```python theme={null}
    from cognis import Cognis

    m = Cognis(owner_id="user_alice")

    ctx = m.get_context(
        messages=[{"role": "user", "content": "What should I do this weekend?"}],
    )
    # ctx["context_string"] — formatted for LLM injection
    # ctx["short_term"] — recent session messages
    # ctx["long_term"] — relevant extracted memories
    print(ctx["context_string"])

    m.close()
    ```
  </Tab>
</Tabs>

## Parameters (Hosted)

| Parameter                 | Type                          | Default  | Description                         |
| ------------------------- | ----------------------------- | -------- | ----------------------------------- |
| `current_messages`        | `List[Dict \| CognisMessage]` | Required | Current turn messages               |
| `owner_id`                | `str`                         | None     | User/tenant scope                   |
| `session_id`              | `str`                         | None     | Session scope                       |
| `agent_id`                | `str`                         | None     | Agent scope                         |
| `max_short_term_messages` | `int`                         | `30`     | Max recent messages to include      |
| `enable_long_term_memory` | `bool`                        | `True`   | Include semantic search results     |
| `cross_session`           | `bool`                        | `False`  | Search memories across all sessions |

## With an LLM

```python theme={null}
import openai
from lyzr import Cognis, CognisMessage

cog = Cognis()

user_msg = "What should I cook tonight?"
context = cog.context(
    current_messages=[CognisMessage(role="user", content=user_msg)],
    owner_id="user_alice",
    session_id="dinner_chat",
    enable_long_term_memory=True,
)

response = openai.chat.completions.create(
    model="gpt-4o",
    messages=[
        {"role": "system", "content": f"You are a helpful assistant.\n\n{context}"},
        {"role": "user", "content": user_msg},
    ],
)
```

## Async

```python theme={null}
context = await cog.acontext(
    current_messages=[CognisMessage(role="user", content="Hello")],
    owner_id="user_alice",
)
```