Metadata-Version: 2.3
Name: karon-mcp
Version: 1.6.0
Summary: Karon API MCP server for Claude, Cursor, and other MCP clients. Retrieve web content from MCP-compatible tools.
Project-URL: Homepage, https://karonlabs.net/api/
Project-URL: Documentation, https://karonlabs.net/api/docs.html
Project-URL: Repository, https://github.com/pplar39/karon-mcp
Author-email: Karon Labs <ceo@karonlabs.net>
License: MIT
Keywords: claude,cursor,mcp,retrieval,web-content
Classifier: Development Status :: 4 - Beta
Classifier: Intended Audience :: Developers
Classifier: License :: OSI Approved :: MIT License
Classifier: Programming Language :: Python :: 3
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: Topic :: Software Development :: Libraries
Requires-Python: >=3.10
Requires-Dist: httpx>=0.27.0
Requires-Dist: mcp>=1.0.0
Description-Content-Type: text/markdown

# karon-mcp

MCP server for the Karon API.

Karon MCP 1.6.0 is the supported public release. Earlier public artifacts are superseded.

Disclosure: `extract` performs model-assisted structured extraction. Page text, prompts, and schemas are processed to produce JSON output. Use `browse`, `scrape`, or `fetch` for non-structured content retrieval.

## Install

```bash
uvx karon-mcp
```

or:

```bash
pip install karon-mcp
```

## Configuration

Set `KARON_API_KEY` in your MCP client environment.

## Tools

Includes 12 MCP tools:
browse, crawl, scrape, fetch, extract, batch_scrape,
watch_snapshot, watch_diff, watch_list, credits, pricing,
browse_with_waf_intel.

`extract` supports `full_access=true` for pages that need the browser-rendered path, and `wait_selector` when a specific element must render before structured extraction.

## Links

- API Documentation: https://karonlabs.net/api/docs.html
- Pricing: https://karonlabs.net/api/pricing.html
