Auto-Enrichment

Redline automatically enriches your nodes with additional data from external sources. This happens in the background so you can keep working while nodes are enhanced.

How It Works

When you create certain node types, Redline queues enrichment jobs that run in the background:

1. You create a node (or the AI creates one)
2. Enrichment jobs are queued based on node type
3. Jobs process in the background (3 at a time, with delays)
4. Node updates automatically with new data

Enrichment Types

Wikipedia Enrichment

Applies to: Actor, Organization

When you create an Actor or Organization node:

• Redline searches Wikipedia for a matching article
• Downloads the summary and thumbnail image
• Populates the description field
• Stores the Wikipedia URL for reference

Example:

Create an Actor node for "Volodymyr Zelenskyy"

Redline fetches:

• Summary from Wikipedia
• Profile photo
• Key biographical details

Article Extraction

Applies to: Newsfeed

When you add a news article URL:

• Extracts the full article text
• Parses author and publish date
• Downloads the featured image
• Generates an AI summary (if enabled)

Example:

Paste a news article URL

Redline extracts:

• Full article content
• Author name
• Publication date
• Featured image

Social Media Enrichment

Applies to: Social

When you create a Social node from a URL:

• Fetches the full post content
• Downloads author profile info
• Captures engagement metrics (likes, reposts)
• Extracts media attachments
• Resolves quoted posts

Supported Platforms:

• Twitter/X
• Bluesky
• Reddit

Authentication Required

Social media enrichment requires you to authenticate with the platform in Settings. Without authentication, only public data is accessible.

OpenGraph Metadata

Applies to: Embed, Social (for links within posts)

For web URLs:

• Fetches OpenGraph metadata (title, description, image)
• Creates a preview card
• Downloads the preview image

Media Archiving

Applies to: All nodes with media

Media files (images, videos, avatars) are:

• Downloaded to local storage for offline access
• Referenced by local path after archiving
• Preserved even if source goes offline

Enrichment Queue

Enrichment jobs are processed with care to avoid overwhelming sources:

Setting	Value
Concurrent jobs	3
Delay between jobs	500ms
Queue check interval	5 seconds
Priority levels	1-10 (lower = higher priority)

Job Priorities

Type	Priority	Description
media_archive	4	Download media files
actor_profile	5	Wikipedia for actors
org_profile	5	Wikipedia for organizations
newsfeed_scrape	5	Article extraction
rssfeed_poll	5	RSS feed polling
social_link_enrich	6	OpenGraph for links

Viewing Enrichment Status

Nodes display enrichment status in the Context Panel:

• Pending - Job queued, waiting to process
• Processing - Currently fetching data
• Complete - Enrichment successful
• Failed - Enrichment failed (may retry)

Manual Re-Enrichment

If enrichment failed or data is stale:

1. Select the node
2. Open Context Panel
3. Click "Re-enrich" or "Refresh"

This re-queues the enrichment jobs.

RSS Feed Polling

RSS Feed nodes have special enrichment behavior:

1. Initial Poll - When created, immediately polls the feed
2. Scheduled Polls - Polls at configured interval (e.g., every 15 minutes)
3. AI Filtering - Each article is evaluated against your relevance prompt
4. Node Creation - Matching articles become Newsfeed nodes on your board

Configurable Settings:

• Poll interval (1 minute to 24 hours)
• Relevance prompt (what topics are you tracking?)
• Relevance threshold (0-100%, how strict?)

Effective Relevance Prompts

Be specific about what you're looking for:

"Articles about Tesla stock price, Elon Musk business decisions, or electric vehicle market trends"

This works better than vague prompts like:

"Interesting technology news"

Performance Considerations

Enrichment runs in the background and shouldn't impact your workflow. However:

• Large imports - Adding many nodes at once queues many jobs
• Rate limits - Some sources may rate-limit requests
• Network issues - Jobs retry on failure but may eventually fail

For large-scale data imports, consider adding nodes in batches.

Privacy Note

Enrichment makes external requests to:

• Wikipedia API
• Target URLs (for scraping)
• Social media APIs (when authenticated)

Your investigation structure (which nodes exist, how they're connected) is never sent externally. Only individual URLs and search queries are made.