Skip to main content
command CX Codex

obsidian-connect — Claude Code command: bridge ideas through your link graph

obsidian-connect — a command that connects two unrelated domains through your Obsidian vault's link graph. Discovers unexpected intersections and generates concrete ideas at their convergence.

MIT ★ 5,412 updated 2026-10-24

Install into Claude Code · Plugin marketplace All methods and platforms ↓
/plugin marketplace add eugeniughelbur/obsidian-second-brain
/plugin install obsidian-second-brain@obsidian-second-brain
Open on GitHub

What obsidian-connect does

obsidian-connect analyzes two domains or topics from your notes and searches for bridges through your link graph. It finds shared links, tags, or concepts; when no direct path exists, it identifies semantic overlap—structural patterns that work across both areas.

The output: 3–5 specific ideas for how a principle from domain A transfers to domain B, what emerges only at their intersection, which opportunities unlock. The command digs for non-obvious connections rather than surface analogies. You can then save the best connections to your ideas folder with links back to both source domains and log the exercise in your daily note.

You need an Obsidian vault with notes linked through [[wikilinks]] and tags. The command looks for _CLAUDE.md in your vault root if it exists, and follows references/ai-first-rules.md when creating new notes.

What it helps with

  • find unexpected connections between separate knowledge domains
  • generate ideas at the intersection of two fields
  • trace paths between concepts through your link graph
  • transfer patterns from one domain to another
  • document and store new connections in your vault

Example prompts for the agent

  • /obsidian-connect "distributed systems" "cooking"
  • /obsidian-connect testing design
  • /obsidian-connect "machine learning" "writing"

How to install obsidian-connect

1Plugin marketplace

/plugin marketplace add eugeniughelbur/obsidian-second-brain
/plugin install obsidian-second-brain@obsidian-second-brain

2Manually: put the files in place

.claude/commands/obsidian-connect.md

Call it with /obsidian-connect. For all projects use ~/.claude/commands/

Questions about obsidian-connect

How do I install obsidian-connect in Claude Code?

Run these commands in order: /plugin marketplace add eugeniughelbur/obsidian-second-brain, then /plugin install obsidian-second-brain@obsidian-second-brain. The command will then be available in Claude Code.

What does obsidian-connect do?

It takes two domains or topics from your notes, searches for connections through your vault's link graph, and generates actionable ideas: how a principle from one domain applies to the other, what concepts exist only at their boundary, and what opportunities open up.

What do I need to set up before using it?

You need an Obsidian vault with notes connected via [[wikilinks]] and tags. If your vault root contains _CLAUDE.md and references/ai-first-rules.md, the command will use them for context and follow their rules when creating new notes.

Source text

obsidian-connect.md · MIT
---
description: Bridge two unrelated domains using your vault's link graph - forces creative friction to spark new ideas
category: thinking
triggers_en: ["connect domains", "cross-pollinate", "bridge ideas", "find an unexpected link"]
triggers_de: ["Themen verbinden", "Ideen kreuzen", "eine Brücke zwischen Ideen schlagen", "eine unerwartete Verbindung finden"]
triggers_es: ["conecta estos dos temas", "cruza ideas de distintos mundos", "busca una conexión inesperada", "conecta esto con algo que no tenga que ver"]
triggers_pt: ["conecte domínios", "cruze ideias", "crie pontes entre ideias", "encontre uma ligação inesperada"]
triggers_zh: ["把这两个领域联系起来", "帮我找跨领域连接", "看看这些想法有什么意外联系", "用我的笔记做跨界联想"]
---

Use the obsidian-second-brain skill. Execute `/obsidian-connect $ARGUMENTS`:

Two arguments required: the two topics, domains, or note names to connect. If only one is given or none, ask the user for both.

1. Read `_CLAUDE.md` first if it exists in the vault root
2. Parse the two domains from arguments (e.g., `/obsidian-connect "distributed systems" "cooking"`)
3. For each domain, search the vault:
   - Find all notes related to that domain (by title, tags, content)
   - Map their backlinks and outgoing links to build a local cluster
4. Find the bridge:
   - Look for shared links, shared tags, or shared people between the two clusters
   - If a direct path exists in the link graph, trace it and explain each hop
   - If no direct path exists, find the closest semantic overlap - concepts, metaphors, or structural similarities
5. Generate creative connections:
   - **Structural analogy**: how a pattern in domain A maps to domain B (e.g., "load balancing is like mise en place - both are about distributing work before the rush")
   - **Transfer opportunities**: what works in A that could be applied to B
   - **Collision ideas**: new concepts that only exist at the intersection of both
6. Present 3-5 specific, actionable connections - not vague analogies but concrete ideas the user could act on
7. Offer to save the best connections to the ideas folder (resolved per `references/folder-map.md` - wiki-style `wiki/concepts/`, Obsidian-style `Ideas/`) with links to both source domains
8. Log the connection exercise in today's daily note

The value is in unexpected links. If the connection is obvious, dig deeper. The best output makes the user say "I never thought of it that way."

---

**AI-first rule:** Every note created or updated by this command MUST follow `references/ai-first-rules.md` - `## For future agent` preamble, rich frontmatter (`type`, `date`, `tags`, `ai-first: true`, plus type-specific fields), recency markers per external claim, mandatory `[[wikilinks]]` for every person/project/concept referenced, sources preserved verbatim with URLs inline, and confidence levels where applicable. If that path does not resolve from your working directory, search upward for it; if you still cannot read it, say so before writing rather than producing a note that silently skips the rule. The vault is for future agent retrieval - not human reading.

**Anti-fabrication:** Search exhaustively before claiming any note, person, or file is absent - false absence is the most common failure mode - and never invent facts, entities, or dates (mark unknowns as `TBD`). See the anti-fabrication and search-completeness hard rules in `references/ai-first-rules.md`.

© Eugeniu Ghelbur · MIT · eugeniughelbur/obsidian-second-brain

Similar slash commands

All slash commands for Notes, tasks and projects →

From the same set51

The whole set →
obsidian-recap obsidian-recap summarizes Obsidian vault activity for today, the week, or the month. It uses parallel subagents to process daily notes, dev logs, and kanban tasks into a structured recap. CX Codex +1 ★ 5,412repo stars obsidian-recurring obsidian-recurring is a command that tracks recurring payments or obligations in Obsidian and calculates the next due date. It links tasks to a board, project note, and log. CX Codex +1 ★ 5,412repo stars obsidian-reindex obsidian-reindex refreshes the semantic search index for an Obsidian vault and reports how many notes are indexed before and after. Runs fast on subsequent calls because it only embeds new or modified notes. CX Codex +1 ★ 5,412repo stars x-pulse x-pulse scans X via Grok x_search for topic trends, popular angles, and post ideas, then saves the research note to your vault. CX Codex +1 ★ 5,412repo stars idea-discovery idea-discovery surfaces 3–5 next-direction candidates by reading ungraduated ideas, open project questions, and orphan research notes in your Obsidian vault. CX Codex +1 ★ 5,412repo stars obsidian-challenge The obsidian-challenge command for Claude Code compares a plan with your Obsidian vault history to find past errors, risks, and contradictions. CX Codex +1 ★ 5,412repo stars obsidian-panel obsidian-panel is a Claude Code command that gathers independent views on a decision and synthesizes them for reviewing disputed questions in an Obsidian vault. CX Codex +1 ★ 5,412repo stars x-read x-read — a command that performs deep analysis of X (Twitter) posts or threads using Grok. Extracts the original text, thread chain, key claims, reply sentiment, and notable voices in the conversation. CX Codex +1 ★ 5,412repo stars

Project sponsors

Companies that keep this analytics open