Connecting Note-Taking Apps

Your Second Brain is only as good as the knowledge it can access. In this lesson, you'll connect your note-taking apps so OpenClaw can index your existing notes and stay in sync as you add new ones.

Prerequisites

Sync Architecture

  Notion / Obsidian         OpenClaw Indexer         Vector Store
  ┌──────────────────┐     ┌──────────────────┐     ┌──────────────┐
  │ Pages & docs     │     │ Chunks text      │     │ Embeddings   │
  │ change over time │────>│ Generates        │────>│ Metadata     │
  │                  │     │  embeddings      │     │ Source links  │
  │ Webhooks or      │     │ Stores vectors   │     │ Timestamps   │
  │  polling detect  │     │ Updates on change│     │              │
  │  changes         │     │                  │     │ Searchable   │
  └──────────────────┘     └──────────────────┘     │ via chat     │
                                                    └──────────────┘

Connect Notion

Notion has a robust API that OpenClaw can use to pull your pages, databases, and their contents.

Go to notion.so/my-integrations and create a new integration
Name it "OpenClaw Second Brain" and select your workspace
Copy the Internal Integration Token
In Notion, share the pages or databases you want indexed with your integration

Add Notion integration

openclaw integrations add notion 
--token "secret_abc123..." 
--sync-mode incremental 
--poll-interval 15m

Share Pages Explicitly

Notion integrations can only access pages you explicitly share with them. Go to each top-level page, click "..." > "Connections" > add your "OpenClaw Second Brain" integration. Child pages inherit access automatically.

Configure which content to index:

notion-sync-config.yaml

notion:
token: "$"
sync:
mode: incremental        # Only process changes since last sync
poll_interval: 15m       # Check for changes every 15 minutes
full_resync: weekly      # Full re-index every Sunday at 2 AMinclude:
type: page
filter: "shared_with_integration"
type: database
ids: ["db-id-1", "db-id-2"]
exclude:
title_contains: "PRIVATE"
title_contains: "Draft"
chunking:
strategy: heading        # Split on headings (H1, H2, H3)
max_chunk_size: 1500     # tokens
overlap: 100             # token overlap between chunks

Connect Obsidian (Alternative)

If you use Obsidian, OpenClaw watches your vault folder for changes. No API key needed — it reads markdown files directly.

Connect Obsidian vault

openclaw integrations add obsidian 
--vault-path ~/Documents/ObsidianVault 
--watch true 
--ignore-patterns ".trash,templates,daily-notes"

obsidian-sync-config.yaml

obsidian:
vault_path: "~/Documents/ObsidianVault"
watch: true                # Real-time file watching
respect_frontmatter: true  # Use YAML frontmatter as metadatainclude:
"**/*.md"
exclude:
".trash/**"
"templates/**"
"daily-notes/**"       # Optional: skip daily journals
metadata_extraction:
tags: true               # Index #tags as metadata
links: true              # Track [[wiki-links]] as connections
frontmatter: true        # Use YAML frontmatter fields

Obsidian Links = Connections

OpenClaw treats Obsidian's [[wiki-links]] as explicit connections between notes. This supercharges the "Discovering Connections" feature in Lesson 5 — the more you link your notes, the smarter your Second Brain becomes.

Run the Initial Index

The first sync indexes all your existing content. This can take a few minutes depending on how many notes you have.

Start initial indexing

openclaw knowledge index --source notion --full

Expected output

Starting full index of Notion workspace...
Found 247 pages across 3 databases
Processing: [████████████████████████████████] 247/247Indexing complete:
Pages processed:    247
Chunks created:     1,832
Embeddings stored:  1,832
Duplicates skipped: 14
Errors:             0
Time elapsed:       2m 34sYour Second Brain now contains 1,832 searchable chunks.
Run 'openclaw knowledge status' to see details.

Verify the Connection

Test that your notes are searchable by asking a question you know the answer to.

Test query

You:  What are my notes about project management?Bot:  I found 12 relevant notes about project management:

"Agile vs Waterfall Comparison" (Notion, updated 3 months ago)
Key point: "Agile works best for teams under 15 people
with unclear requirements..."


"Meeting Notes: PM Strategy Session" (Notion, 6 months ago)
Key point: "We decided to adopt shape-up methodology
for the Q3 cycle..."


"Book Notes: The Mythical Man Month" (Obsidian, 1 year ago)
Key point: "Adding people to a late project makes it later..."

Want me to go deeper on any of these?

If you have a massive note collection, the initial index can take a while. Optimize with:

Performance tuning

indexing:
batch_size: 50             # Process 50 pages at a time
concurrency: 4             # 4 parallel embedding requests
rate_limit: 100            # Max 100 API calls per minute
priority_index:            # Index these first
"Work/**"
"Projects/**"
defer_index:               # Index these last
"Archive/**"

For ongoing sync, incremental mode only processes changed files, so it's fast regardless of vault size.

Many people use Notion for work and Obsidian for personal notes. OpenClaw handles multiple sources seamlessly — the vector store doesn't care where the content came from. Just add both integrations and queries will search across both.

You can filter by source in queries: "What do my Notion notes say about X?" vs "Search only my Obsidian vault for Y."

Notion Integration Docs

Complete API reference for Notion sync configuration, including database filters and property mapping.

https://docs.openclaw.ai/integrations/notion

Knowledge Check

Why does OpenClaw use 'chunking' when indexing your notes?