Convert PDF, Word (.docx), PowerPoint (.pptx), Excel (.xlsx) and images to Markdown — headings, tables, lists and embedded images intact. Parsed natively: no Python, no pip, no cloud service, nothing uploaded. Local OCR reads text in images.
A real terminal with shell integration and beginner-friendly command help, plus a multi-agent workflow for Claude Code sessions: word-level diffs, undo, backlink-breakage warnings, and more.
Manage AI skills, agents, commands, and rules for Claude Code, Cursor, Codex, and more, with usage insights and MCP visibility.
An independent bookmarks sidebar that opens saved links in Obsidian's built-in Web Viewer, so you never have to leave the app to browse.
Group open tabs into named, switchable spaces you can tuck away and complete when the work is done.
Preview PowerPoint and Word documents inside Obsidian. Automatically refreshes when you, Claude, or another AI agent makes changes to the file, so your preview is always up to date. Renders via a local LibreOffice install (bring your own, or install one from Settings).
Reveal hidden dot-folders (like .claude or .github) in the file explorer, search, and Bases, with live updates, file-type filters, and exclusions.
Embed any website as an interactive browser view inside a note.
Embed, create, and sync Asana tasks inside Obsidian. Supports any workspace or project.
Ask a research question about your vault and get an answer with grounded, claim-grouped citations -- no external account, uses your local claude CLI.
Publishes your vault as a static website, rendered by Obsidian itself, so Dataview, Bases and Mermaid output survives.