We don't bundle an AI model. We bundle the place your AI output goes. And we expose a REST API so the rest of your stack can put docs in and take them out.
If it outputs Markdown, we render it. The list below is the one we explicitly test against — every common AI assistant fits here.
Paste from the web app, the desktop app, the macOS quick-action — formatting survives.
Paste from Claude.ai or any tool wrapping Claude. Tables, lists, code — all survive.
Paste from Gemini directly or via the Workspace AI tools.
Paste from M365 Copilot or GitHub Copilot Chat. Markdown stays Markdown.
Pasted answers keep their footnoted links intact.
Cursor. Cline. Windsurf. A pasted Slack export. Whatever you use this year.
Programmatic doc + folder management for Zapier, n8n, and your own scripts.
Triggers and actions — create docs from form submissions, push meeting notes from your calendar, archive Slack threads as searchable docs.
Same API. Same auth. Wire sweetdocs into your homelab.
10K calls / month on Pro, 100K on Teams. Personal and org scopes. Read the docs on /api.
API access is included on every paid plan. Read the API docs →