{{ selectedSite.site_name }}
{{ currentRules.input_mode || 'url' }}Lower numbers take precedence over higher numbers.
Defines the primary data format for this source.
| Name | Value | Domain | × |
|---|---|---|---|
| No cookies configured | |||
Specifically for PDF and Image content. Will use Gemini Vision or local OCR engine to textify content.
Skip mode: no textification
Retaining physical format for raw LLM processing
Storage / Preview Strategy (L3)
Determines how content is persisted for long-term reference{{ strategy.desc }}
Snapshot Storage Notice
Snapshots will store the full HTML content in the database. This significantly increases database size. Ensure the rendering node is correctly configured for high-fidelity capture.
Automatic provider detection is used when possible.
Records the source URL for reference (no HTML content will be fetched or stored).
Testing moved to Pipeline Tab
Use the full Pipeline Test for extraction + AI analysis