An Obsidian vault is a folder full of Markdown files, which is exactly why there isn’t one way to connect it to Claude, but three. All three work. They differ in what you have to put in and what comes out the other end. Here they are side by side, each with its limits, and the table at the bottom.
Route 1: Share the vault as a folder
The most direct route, and it costs nothing: start Claude Code in the vault folder (or share the folder with Claude Desktop) and drop in a CLAUDE.md that explains the structure. Claude then reads and searches the files like any other project.
What it does well: For text notes on one machine this is hard to beat. No setup, no cost, and data only leaves your machine for the model request itself. You version changes with Git.
Where it ends: The access exists only on that machine: on your phone and in claude.ai the folder doesn’t exist. Claude searches with file tools (grep), not semantically: it finds “notice period” when the text says “notice period”, but not when the note talks about “contract duration”. And a scan in the vault stays a file name.
Obsidian MCP: local server or hosted
Routes 2 and 3 both run on MCP, the open standard AI apps use to talk to outside services. The difference is where the MCP server runs: on your machine, right next to the vault, or with a provider that holds a copy of your notes. Obsidian has no official MCP server (as of November 2026). People have been asking for one in the Obsidian forum since December 2025; everything available comes from the community.
Route 2: A local MCP server
The community maintains several Obsidian MCP servers. Some you start separately (usually via npx, configured with your vault path), others run as a community plugin inside Obsidian. Claude gets real tools (read note, create note, search vault) instead of raw file access.
What it does well: A cleaner interface than route 1, sometimes with better search. Creating and editing notes is structured. Everything stays local.
Where it ends: It remains DIY: install Node or a plugin, maintain a config file, track updates. You can’t tell in advance which project will still be maintained a year from now, and Obsidian hasn’t reviewed them. The server runs while your machine runs. It answers neither the multi-device question nor the scan question.
Route 3: The hosted knowledge graph
Import the vault into Memexa as a ZIP (wikilinks survive), then connect Claude via MCP (the three-minute guide is here). From then on every AI app reads the same state: Claude Desktop, claude.ai, Claude Code, and ChatGPT or Cursor too.
What it does well: Hybrid search (semantic + full-text) instead of grep. Documents and scans go through text recognition and become citable, down to the page. Works on every device without anything running on your machine.
Where it ends: It’s a hosted service: your data lives with a provider (in our case: EU, GDPR, Markdown export any time), and it costs money: from €2.49 a month billed yearly or €2.99 monthly, with no free plan; the 14-day trial requires a card (pricing). If you work strictly “nothing leaves my machine”, take route 1 or 2.
The table
| Vault folder | Local MCP server | Hosted graph (Memexa) | |
|---|---|---|---|
| Setup effort | Minutes | Node + config upkeep | Minutes (URL + sign-in) |
| Multiple devices / claude.ai | No | No | Yes |
| Documents with scans | No | No | Yes (OCR) |
| Page-level citations | No | No | Yes |
| Where the data lives | Your machine | Your machine | EU servers |
| Cost | €0 | €0 | from €2.49/mo (yearly) |
The honest reading: on “where the data lives” and on cost, DIY wins, and if all you need is your own text notes on one machine, route 1 finishes the job. The two rows where the hosted route stands alone (scans and citations) only matter if your knowledge includes documents. But then they decide everything.
Obsidian AI without plugin tinkering
Obsidian itself ships no AI. Search for “Obsidian AI” and you end up with community plugins: Smart Connections finds related notes through semantic search and uses a local model for that by default; Copilot for Obsidian adds a chat to the sidebar, with your own API key (OpenAI, Anthropic and others) or a local model. Both plugins are free; you pay the provider for API usage.
That works, with two catches. You maintain the plugin, the model choice and the keys yourself. And the AI lives inside the Obsidian window: in Claude or ChatGPT, where you may do most of your work anyway, nobody knows your vault exists.
The reverse route needs no plugin: instead of bringing AI into Obsidian, you bring your notes to the AI app you already use. Route 1 needs only the folder and Claude Code; route 3 needs only the vault as a ZIP and an address you add in Claude or ChatGPT. The cost of route 3: the import runs periodically, not live, so edits in Obsidian arrive with the next import.
Frequently asked questions
Is there an Obsidian AI?
Not built in (as of November 2026). Obsidian has neither its own AI assistant nor an official MCP server. AI features come through community plugins such as Smart Connections or Copilot for Obsidian, or you connect your notes to Claude or ChatGPT via MCP, locally (route 2) or hosted (route 3).
Do I have to give up Obsidian?
No. We recommend a division of labor: Obsidian stays the writing tool, the Memexa import runs periodically, and Claude works on the graph. Wikilinks survive the import as real edges.
What happens to my plugins and templates?
They stay in Obsidian and keep working: the import reads the Markdown files, not the Obsidian configuration. Dataview queries and similar plugin syntax arrive as text, not as executed queries.
Does this work with ChatGPT instead of Claude?
Yes. Routes 2 and 3 speak MCP, which ChatGPT supports too. Route 1 is Claude-specific (folder sharing + CLAUDE.md).