Metadata-Version: 2.5
Name: langchain-capturekit
Version: 0.1.0
Summary: LangChain tools for CaptureKit: read web pages as clean Markdown and get an AI profile of a website.
Project-URL: Homepage, https://www.capturekit.dev
Project-URL: Documentation, https://github.com/capturekitdev/langchain-capturekit#readme
Project-URL: Repository, https://github.com/capturekitdev/langchain-capturekit
Project-URL: Issues, https://github.com/capturekitdev/langchain-capturekit/issues
Author-email: CaptureKit <contact@capturekit.dev>
License-Expression: MIT
License-File: LICENSE
Keywords: agents,capturekit,content-extraction,langchain,markdown,tools,web-scraping,website-analysis
Classifier: Intended Audience :: Developers
Classifier: Operating System :: OS Independent
Classifier: Programming Language :: Python :: 3 :: Only
Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Programming Language :: Python :: 3.13
Classifier: Topic :: Internet :: WWW/HTTP
Classifier: Typing :: Typed
Requires-Python: >=3.10
Requires-Dist: httpx<1,>=0.27
Requires-Dist: langchain-core<2,>=0.3.15
Provides-Extra: test
Requires-Dist: langchain-tests>=0.3.5; extra == 'test'
Requires-Dist: pytest-asyncio>=0.24; extra == 'test'
Requires-Dist: pytest>=8.3; extra == 'test'
Requires-Dist: ruff<1,>=0.8; extra == 'test'
Description-Content-Type: text/markdown

# langchain-capturekit

[LangChain](https://www.langchain.com) integration for [CaptureKit](https://www.capturekit.dev): read web pages as clean Markdown and get an AI profile of a website.

## Installation

```bash
pip install -U langchain-capturekit
export CAPTUREKIT_API_KEY="your-api-key"
```

Get an API key at [app.capturekit.dev](https://app.capturekit.dev).

## Tools

| Class | Tool name | What it does |
| --- | --- | --- |
| `CaptureKitPageContent` | `capturekit_page_content` | Load a web page in a real browser through CaptureKit and return it as Markdown. |
| `CaptureKitAnalyzePage` | `capturekit_analyze_page` | Get an AI profile of a web page through CaptureKit. |

Each tool takes `api_key` (defaults to `CAPTUREKIT_API_KEY`), `base_url` and `timeout`, plus the settings shown in its docstring. Errors (invalid key, no credits, rate limit, timeout) come back to the model as a `ToolException` message, without the API key. Every tool has an async version.

```python
from langchain_capturekit import CaptureKitPageContent

tool = CaptureKitPageContent()
print(tool.invoke({"url": "https://example.com"}))
```

### In an agent

```python
from langchain.agents import create_agent
from langchain_capturekit import CaptureKitPageContent, CaptureKitAnalyzePage

agent = create_agent("openai:gpt-4.1-mini", tools=[CaptureKitPageContent(), CaptureKitAnalyzePage()])
result = agent.invoke({"messages": [{"role": "user", "content": "..."}]})
```

### Document loader

`CaptureKitLoader` loads pages as Markdown `Document`s, with `source`, `title` and `description` in the metadata:

```python
from langchain_capturekit import CaptureKitLoader

docs = CaptureKitLoader(["https://example.com", "https://example.org"], continue_on_failure=True).load()
```

## Development

```bash
pip install -e ".[test]"
pytest tests/unit_tests
```

## License

MIT
