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

# Python SDK (ksapi)

> Install, authenticate, and make your first API call in under five minutes.

[![PyPI](https://img.shields.io/pypi/v/ksapi)](https://pypi.org/project/ksapi/) [![GitHub](https://img.shields.io/badge/GitHub-ks--sdk--python-181717?logo=github)](https://github.com/knowledgestack/ks-sdk-python)

## Install

<CodeGroup>
  ```bash pip theme={null}
  pip install ksapi
  ```

  ```bash uv theme={null}
  uv pip install ksapi
  ```

  ```bash poetry theme={null}
  poetry add ksapi
  ```
</CodeGroup>

Requires Python ≥ 3.9.

## Authenticate

<Steps>
  <Step title="Sign up">
    Create an account at [app.knowledgestack.ai](https://app.knowledgestack.ai).
  </Step>

  <Step title="Issue an API key">
    From the dashboard, create an API key scoped to your user. It looks like `sk-user-...`.
  </Step>

  <Step title="Export it">
    ```bash theme={null}
    export KS_API_KEY="sk-user-..."
    export KS_BASE_URL="https://api.knowledgestack.ai"   # optional; prod is default
    ```
  </Step>
</Steps>

## First call

```python theme={null}
import os
import ksapi
from ksapi.api.folders_api import FoldersApi

config = ksapi.Configuration(
    host=os.environ.get("KS_BASE_URL", "https://api.knowledgestack.ai"),
    access_token=os.environ["KS_API_KEY"],
)

with ksapi.ApiClient(config) as client:
    folders = FoldersApi(client)
    for folder in folders.list_folders():
        print(folder.id, folder.name)
```

## Read a document

```python theme={null}
from ksapi.api.path_parts_api import PathPartsApi
from ksapi.api.documents_api import DocumentsApi

with ksapi.ApiClient(config) as client:
    hits = PathPartsApi(client).find_path_parts(query="credit policy")
    doc_id = hits[0].path_part_id
    doc = DocumentsApi(client).read_document(path_part_id=doc_id)
    print(doc.text)
```

## Building an agent?

<Warning>
  For agent / RAG workflows, use the [MCP server](/sdks/mcp-server) instead. It plugs into every major agent framework without writing glue code.
</Warning>

## Reference

The full auto-generated API reference lives in the [ks-sdk-python repo](https://github.com/knowledgestack/ks-sdk-python/tree/main/docs). It's also mirrored under `sdks/_generated-python/` in this docs repo, refreshed nightly.