Document Collection

Revornix supports four document entry types: website documents, quick notes, file documents, and audio documents.

The Jina engine currently supports only website documents. If you need file documents, switch the conversion engine to something other than Jina.

You can collect content directly in the product UI, or through APIs, SDKs, extensions, and automation tooling.

Web collection

Configuration

Link: the website URL you want to ingest.
Tags: document tags used for later filtering and organization.
Auto tag: the system selects tags from your tag list based on document content.
Auto summary: the system uses the default document-reading / summary model to generate summary fields and replace the default title and description.
Auto podcast: the system generates an audio podcast from the document and attaches it to the document.
Push to sections: select sections during creation so the document enters those section workflows immediately.

Note: if auto summary is enabled, the title and description you set will not take effect and will be replaced by the AI summary.

API collection

To help developers integrate third-party apps or build custom workflows, Revornix provides API-based ingestion.

Click the user avatar in the bottom-left, choose Account, then go to APIKey Management and click Configure on the right.

For request parameters and response examples, see Third-Party API.

Chrome extension collection

See the plugin repository: Revornix-Chrome-Extension

JS package collection

See the NPM package: Revornix-Npm-Lib

Python library collection

If you want to ingest content from Python scripts, backend jobs, or CI workflows, start with Python SDK and CLI.

Source and installation details: Revornix-Python-Lib

OpenClaw skill collection

If you build automation agents in OpenClaw, you can use the dedicated OpenClaw Skill so the agent can call Revornix document, section, and label operations directly.

File and audio notes

File documents depend on the default file parsing engine.
Audio documents depend on the default transcription engine and are uploaded into the default file system before transcription starts.
The web upload flow accepts common document and presentation formats, including pdf / doc / docx / ppt / pptx / jpg / jpeg / png.