Leviathan Systems SDK documentation

JavaScript SDK

Install the versioned npm-compatible archive from this website. It includes the client and the command-line tool, with no install scripts or third-party runtime dependencies. Use Node.js 20 or later; the client module also works in modern browsers on the same origin.

npm install https://www.leviathansystems.co/downloads/leviathan-public-content-1.0.0.tgz

import { LeviathanClient } from '@leviathansystems/public-content';
const client = new LeviathanClient();
const page = await client.listResources({ kind: 'service', limit: 10 });
const resource = await client.readResource(page.items[0].id);
console.log(resource.canonical, resource.markdown);

Download JavaScript package 1.0.0 · Read the client source

Python SDK

Use Python 3.10 or later. The pure Python wheel uses the standard library and has no runtime dependencies.

python -m pip install https://www.leviathansystems.co/downloads/leviathan_public_content-1.0.0-py3-none-any.whl

from leviathan_public_content import LeviathanClient
client = LeviathanClient()
page = client.list_resources(kind='service', limit=10)
resource = client.read_resource(page['items'][0]['id'])
print(resource['canonical'])

Download Python wheel 1.0.0 · Read the Python source

These packages are distributed directly from the official website. They are not npm or PyPI registry listings.

Pagination and recovery

JavaScript resourcePages(options) is an async iterator; Python resource_pages(**options) is a generator. Both preserve the query and origin while following next_cursor. A cursor is tied to a content revision; begin a new search when it expires.

LeviathanAPIError exposes the HTTP status, stable error code and retry interval. JavaScript uses retryAfter; Python uses retry_after. The clients do not automatically retry. Honor 429 and Retry-After, correct invalid queries, and keep retrieval sequential. A resource may have a null Markdown body; follow its canonical HTML URL in that case.