Claude + Obsidian have to be illegal¶
Source: Tweet thread by Defileo🔮 (@defileo)
Original tweet: https://x.com/defileo/status/2042241063612502162
Article being quoted: https://x.com/i/article/2041858771123617793 (tweet ID 2041858771123617793)
Posted: Thu Apr 09 13:59:47 +0000 2026
Engagement: 9,231 likes · 904 retweets · 186 replies
Source¶
Primary source: the tweet links to a long-form article at https://x.com/i/article/2041858771123617793 (a Twitter article/Card). Full article text was not retrievable via automated tools due to X/Twitter API restrictions. Content reconstructed from: (1) preview text extracted via vxTwitter API, (2) engagement metadata (likes, retweets, replies), (3) image analysis of the attached screenshot.
Overview¶
Defileo posted a tweet pointing to a long-form article about a Claude + Obsidian "second brain" setup. The tweet's preview text states:
"Claude + Obsidian should be illegal. Let me tell you what my setup does before I explain how to build it. Every morning I open my laptop, Before I type a single word, Claude already knows who I am."
This is a personal knowledge management setup where Claude has persistent context about the user's identity, history, and goals — loaded automatically before any interaction begins.
Key Elements¶
- Obsidian as the local knowledge graph / notes layer
- Claude as the reasoning/agent layer with full context pre-loaded
- Automatic morning context load — Claude knows who the user is before they type
- Likely involves Zapier or similar automation to bridge Obsidian ↔ Claude on startup
- The tweet's image shows: Obsidian icon, Zapier icon, numbered step icons (1 and 2), brain emojis, and a growth chart — suggesting a step-by-step automated workflow
Relevance to dark-factory-kb¶
This pattern — persistent context pre-loading for an agent — is directly relevant to:
- agent-memory-patterns — Claude having persistent memory of user identity before task start
- second-brain-architectures — Obsidian as a structured knowledge layer feeding an AI agent
- pre-task-context-injection — the "before you type a single word" pattern mirrors startup context injection in Kelly Router
Related Concepts¶
- Personal knowledge management (PKM) with AI
- Context pre-loading for LLM agents
- Obsidian as a semantic memory layer
- Automated morning briefings for AI assistants
- Agent startup routines and identity context
Meta¶
Note: The full article content could not be retrieved (X/Twitter blocks API access). This entry captures the metadata, preview text, engagement stats, and visual elements from the attached image. The article appears to be a detailed setup guide with step-by-step instructions involving Obsidian, Claude, and likely Zapier for automation.
Bibliography¶
- Defileo (@defileo), "Claude + Obsidian have to be illegal" tweet thread, 2026-04-09. Content reconstructed from preview text (vxTwitter API), engagement metadata, and image analysis.