X Bookmark Absorb

X Bookmark Deep Absorption

Transform raw X bookmark data into full wiki graph integration — not just hub-page entries.

The Problem This Solves

Shallow absorption: bookmarks get added to x-bookmarks-*.md hub pages but never propagate to dedicated tool/person pages, cross-references, resolvers, or the index. The wiki stays disconnected from the signal.

When to Run

• After ft sync + ./scripts/sync-x-bookmarks.sh adds new bookmarks
• When Kevin says "absorb bookmarks" or "process bookmarks"
• During source-compile automation runs targeting x-bookmarks
• When Kevin says bookmarks from a date range need processing

The Pipeline

Phase 1: Extract and Categorize

# Parse bookmarks.jsonl for the target date range
# Categorize each bookmark:
#   TOOL — new library, CLI, framework, skill, platform
#   DESIGN — typography, color, animation, layout, UI pattern
#   PERSON — notable individual (2+ mentions or high-signal take)
#   PHILOSOPHY — strong opinion, thesis, counter-signal
#   AI — agent tools, model announcements, agent UX
#   OTHER — lifestyle, ephemeral, low-signal

For each bookmark, extract: date, author handle+name, full text, links, engagement (likes, bookmarks), media type.

Phase 2: Hub Page Entries (Level 1)

Route each bookmark to the correct hub page:

Format: ### Title block with quoted tweet, author attribution, date, engagement, and a **Takeaway:** line linking to relevant wiki pages.

Phase 3: Dedicated Pages (Level 2)

A bookmark gets a dedicated page when ANY of these are true:

• Tool/library announced with >500 bookmarks or a working demo
• Person page warranted — notable individual mentioned 2+ times across sources OR single take with >5K likes from a builder with credibility
• Concept/philosophy — original thesis worth tracking over time
• Existing wiki page should be updated — the bookmark adds new info to an entity already in the wiki

Page creation follows wiki/RESOLVER.md decision tree. Use the standard page format from AGENTS.md.

Phase 4: Cross-Reference Walk (Level 3)

For every dedicated page created or existing page updated:

1. Search wiki for related pages — rg for the entity name, author handle, related concepts
2. Add timeline entries to related pages that should know about this
3. Update frontmatter related: arrays on both the new and existing pages
4. Add [[wikilinks]] in body text where the entity is mentioned

Phase 5: Registry Updates (Level 4)

If any bookmark introduces a tool that could enter Kevin's stack:

1. Check if it belongs in config/cursor/rules/tool-hierarchy.mdc
2. Check if it belongs in wiki/SKILL-RESOLVER.md
3. If it's a skill candidate, note it for evaluation (Brin + skill-auditor + reputation)

Phase 6: Index and Log (Level 5)

1. Add new pages to wiki/_index.md (alphabetical within category, update counts)
2. Append to wiki/log.md with full absorption record
3. Run npx tsx scripts/build-index.ts if available

Notability Gates

Gets a dedicated page:

• Tool/skill with >500 bookmarks or a GitHub repo with >100 stars
• Person with >10K followers AND a take >5K likes relevant to Kevin's domains
• Any tool Kevin's own tweet references (self-signal)
• Any entity mentioned across 2+ separate bookmarks

Hub-page only (no dedicated page):

• Ephemeral deals/promos (ChatGPT free months, etc.)
• Lifestyle/personal content (NYC photos, food, etc.)
• Low-engagement takes (<200 likes) from non-notable accounts
• Link-only tweets with no extractable content

Skip entirely:

• Retweets of content already captured
• Duplicate topics already well-covered in existing hub entries

Quality Checklist

Before declaring absorption complete:

• Every bookmark from the date range is in at least one hub page
• Every notable bookmark has a dedicated page or existing page update
• Every new page is in _index.md with correct category and count
• Every person page has proper aliases (handle, @handle, full name)
• Cross-references flow BOTH directions (new→existing AND existing→new)
• log.md records what was created, updated, and why
• No orphan pages (every new page has at least one inbound link)