absorb

Pre-Absorb Check

Run: python3 ${CLAUDE_PLUGIN_ROOT}/skills/verify/scripts/verify.py {vault}/knowledge/raw --status

• ❌ (false): BLOCK — "Run /verify first before absorbing."
• 👁️ (manual): Run WITH approval prompt — show each proposal, ask "Approve? (all / pick numbers / skip)"
• ✅ (true): Run autonomously, bypass approval.

Vault Location

Vault path: Read from ${CLAUDE_PLUGIN_ROOT}/brain-os.config.md

Behavior

If /absorb [book-name] — Deep dive on one book

1. Read the book note from {vault}/knowledge/books/[book].md
2. Read your current context — {vault}/context/strategy.md, {vault}/context/goals.md, {vault}/context/business.md
3. Cross-reference book insights against your actual situation
4. Generate absorption report with 3 sections:

Section 1: Business Applications

• How does this book's insights apply to [[context/business]]?
• What should change in [[context/strategy]] based on this?
• Any implications for [[context/icp]] or [[context/brand]]?

Section 2: Personal Applications

• What habits, mindsets, or practices should I adopt?
• How does this change how I think about [[context/goals]]?

Section 3: Thinking Seeds

• What new ideas does this book seed?
• What existing ideas in {vault}/thinking/ideas/ does this reinforce or challenge?
• What connections to other books or knowledge emerge?

5. Propose vault updates:
   • "Should I add [insight] to your strategy doc?"
   • "Should I create a new idea note in thinking/ideas/?"
   • "Should I update your preferences/rules based on [learning]?"
6. Wait for approval before making any changes

If /absorb (no argument) — Cross-book synthesis

1. Read all recent book notes (last 5-10 books)
2. Find cross-book patterns — themes that appear across multiple books
3. Identify knowledge gaps — topics you keep reading about but haven't crystallized into action
4. Generate synthesis report:
   • Common threads across your reading
   • Contradictions between authors
   • The strongest actionable insights (backed by multiple sources)
   • Suggested {vault}/thinking/patterns/ entries for recurring themes
5. Save report to {vault}/thinking/agent-output/YYYY-MM-DD-absorption-synthesis.md

Output Format (Single Book)

## Absorption Report: [Book Title]

### For Your Business (AI Leaders Vietnam)
| Insight | Applies to | Suggested Action |
|---------|-----------|-----------------|
| [Insight] | [[context/strategy]] | [What to do] |
| [Insight] | [[context/icp]] | [What to do] |

### For You Personally
- [How this changes your thinking or habits]

### Thinking Seeds
- **New idea**: [Idea name] — [Brief description]
  → Create in thinking/ideas/? [y/n]
- **Reinforces**: [[thinking/ideas/existing-idea]] — [How]
- **Challenges**: [[thinking/patterns/existing-pattern]] — [How]

### Vault Updates Proposed
1. Add to context/strategy.md: [what]
2. Create thinking/ideas/[new-idea].md
3. Update context/preferences.md: [new rule from book]

Approve updates? (all / pick numbers / skip)

Rules

• Never update vault files without explicit approval
• Always show what will change before changing it
• Prioritize actionable insights over theoretical knowledge
• Connect to EXISTING vault content — don't create orphan notes

Outcome log

Follow skill-spec.md § 11. Append to {vault}/daily/skill-outcomes/absorb.log:

{date} | absorb | {action} | ~/work/brain-os-plugin | {output_path} | commit:{hash} | {result}

• action: absorb (single book) or synthesis (cross-book)
• output_path: path to absorption report or synthesis file
• result: pass if vault updates applied, partial if user skipped some proposals, fail if blocked by verify
• Optional: args="{book-name}", corrections=N