Capture Methods

Capture is the cheapest, fastest path from a thought in your head to a file in your vault. MindWiki gives you several ways in, all of which route to the same place: capture/ in your vault.

The capture inbox

Every account has one pre-built folder: capture/. Anything captured through any method below lands here as a markdown file. You can let it sit, or move it elsewhere, or link to it from other pages — your choice.

The web app dashboard shows a Capture Inbox panel with the most recent items. The macOS app shows the same folder in the sidebar.

1. Email capture

Every account gets a personal capture address based on your username:

{username}@mindwiki.io

Forward any email to that address — newsletters, articles, notes from yourself, threads you want to remember — and it lands in your capture/ folder as a parsed page. The subject becomes the title, the body becomes the page content, and metadata like sender, date, and source URL is preserved.

You'll find your exact email address in two places:

• The web account dashboard at mindwiki.io/account
• The Settings page at mindwiki.io/account/settings

You can copy it to your clipboard with one click and add it as a contact in your mail client for one-tap forwards.

Email capture is on by default for every account. You don't need to set anything up. Attachments are noted in the captured page but are not yet stored as separate files.

2. Web Quick Capture

The web app has a Quick Capture surface visible from the account dashboard at mindwiki.io/account. Type or paste a thought, click Capture, and the entry lands in capture/ as a fresh markdown file.

This is the fastest method when you're already on the web. No file naming, no folder picking — just write and ship.

3. In-app page creation

When you want to skip capture/ entirely and start a page in a specific folder, both clients support direct creation:

• macOS app — Cmd+N creates a new page wherever the file tree is currently focused. Cmd+Shift+N creates a new folder.
• Web app — open the command palette with Cmd+K and pick New page or New folder.

This is the right method for content that already has a clear home — meeting notes that belong in projects/foo/, a reference that belongs in reference/, etc.

4. MCP capture from your AI

Any AI client connected through MCP can call the mindwiki_capture tool to drop content directly into your capture/ folder. For example, mid-conversation in Claude or ChatGPT, you can say:

Capture this to my MindWiki vault: <whatever you want saved>

The AI calls mindwiki_capture, MindWiki creates a new dated page in capture/, and the captured item shows up across all your devices within seconds. This is especially useful for saving useful AI outputs without copy-pasting.

The mindwiki_capture tool requires the write scope on the connected token. OAuth connections have full read+write by default. API keys obey whatever scopes you selected at creation. See API Access.

5. REST API capture

For scripts, automations, and integrations with tools like Readwise, Notion, ClickUp, Zapier, or Make, mint a personal API key at mindwiki.io/account/api-keys and POST to the capture endpoint:

curl -X POST "https://api.mindwiki.io/vault/capture" \
  -H "Authorization: Bearer $MINDWIKI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"title":"Readwise highlight","source":"readwise","content":"A saved highlight goes here."}'

The endpoint generates a path like capture/2026-05-09-readwise-highlight.md automatically. Required parameters: content. Optional: title, source. Requires a key with write scope.

See API Access for the full integration guide.

What lives in a captured page

Every captured page is a normal markdown file. You can edit it, link to it, move it, rename it, or delete it like any other page. There's no special "captured" state — capture/ is just a folder, and everything inside it is just markdown.

When you forward an email, the page typically includes:

• Title from the email subject
• Frontmatter with source: email, sender, and timestamp
• The email body in markdown
• Original sender and recipient information

When you capture from MCP, the page includes:

• Title (auto-generated from content if you don't supply one)
• Frontmatter with source (often the AI client name) and timestamp
• Your captured content as the body

Capture method availability

Where to go next

• Vault Structure — what to do with captures over time
• The Editor — turning a capture into a real page
• API Access — programmatic capture from your own tools