MODSetter

SurfSense: Research and Edit Documents on Your Desktop

SurfSense is a privacy-focused desktop AI agent for researching, transforming, and editing documents that you cannot upload to a hosted service. It keeps its index on your machine, supports Windows, macOS, and Linux, and lets you choose a local or remote model.

ActivePythonApache-2.0AI AgentsDesktop AppPrivacy
Stars
16.6k
Forks
1.6k
Stars, last 30 days
+372
Last commit
3 days ago
Contributors
100+

Our take

Activeregular commits and releases

Health score

75/100

How it is scored
Activity
84
Community
85
Issues
46
Pull requests
84

+372 stars in the last 30 days

  • Merges more pull requests than 96% of the projects we track
  • Releases more often than 84% of the projects we track
  • More contributors than 78% of the projects we track

SurfSense is an actively maintained desktop option for private document research and editing, with frequent releases and contributions spread across a broad contributor base.

Good fit if

  • You need to research or edit files that should remain on your machine, and can use a local model when remote processing is not acceptable.
  • You want a desktop workflow across Windows, macOS, or Linux, with document editing as well as question answering.
  • You want to inspect, adapt, or redistribute the software under Apache-2.0 terms.

Look elsewhere if

  • You need a hosted, account-based collaboration workflow rather than a desktop app.
  • Your workflow depends on a specific model or service that SurfSense does not support through its local or OpenAI-compatible API options.
  • You need every format feature to work without optional dependencies: Office-to-PDF conversion requires LibreOffice, and image generation requires an image model.
All health signals
Last commit2026-10-21 (3 days ago)
Commits, last 90 days100+
Releases, last 12 months40 (latest v2.2.3, 2026-10-21)
Contributors100+ (top contributor: 37% of commits)
Issues closed, last 90 days1+
Pull requests merged, last 90 days95 (typically merged in 0 days)
Project age2 years

Checked on 2026-10-25 with the GitHub API.

Project history

Daily GitHub snapshots since Oct 7, 2026. History covers stars and forks; health signals are not tracked historically.

Selected snapshot

Stars

16,893

Forks

1,613

Change since Nov 17, 2026: +3 stars·+1 forks

Overview

SurfSense is a free, Apache-2.0-licensed desktop app for using an AI agent to research, transform, and edit documents on your own machine. It is aimed at people who want a NotebookLM-style workflow for sensitive files they cannot upload to a hosted service. The index stays on disk, and the app does not require an account.

The app runs on Windows, macOS, and Linux. You choose the model, either local or remote; using a remote model means your requests and relevant document content may leave your machine. SurfSense supports research with cited answers across formats including PDF, Word, Excel, PowerPoint, images, audio, and text, as well as document creation and editing.

Key Features

  • Index local files and folders, then ask the agent to find information or extract answers with citations.
  • Research PDFs, including scanned pages, and read spreadsheet formulas and audio transcripts.
  • Create and edit documents, spreadsheets, presentations, and PDFs. File edits and conversions produce a new file rather than changing the original.
  • Transform supported documents, for example by merging or splitting PDFs, converting Office files to PDF, or analyzing CSV data.
  • Generate mind maps, flashcards, quizzes, and podcast audio from a transcript.
  • Extend research through plugins for web search, site crawling, YouTube, and other sources, plus optional third-party MCP servers.
  • Choose a local model or a remote model through an OpenAI-compatible API.

Use Cases

  • A researcher can search a collection of local PDFs and get cited answers without uploading the documents to a hosted notebook service.
  • An analyst can ask questions about Excel workbooks, including formulas, or explore CSV data and create charts.
  • A team member can revise a Word document or presentation while keeping the original unchanged.
  • A learner can turn local notes and recordings into study aids such as flashcards, quizzes, or mind maps.
  • A user can add web research to a local-document workflow when the source material needs to include public web pages or videos.

What you need

Detected in the repository

  • Python (main language on GitHub)
  • Automated checks on GitHub Actions

License in plain words

Apache-2.0permissive

  • Commercial use: yes
  • Modify and redistribute: yes
  • You must keep: the license, the NOTICE file and a note of your changes
  • Share your changes: no
  • Includes an explicit patent grant from the contributors:

A summary, not legal advice: the LICENSE file is what applies.

Getting Started

Download the installer for your platform from the SurfSense downloads page, complete onboarding, then add files or folders as sources. See the README for format details, plugins, and setup guidance.

Alternatives

  • local-deep-research: Local Deep Research focuses on cited research reports across web, academic, and private sources, while SurfSense emphasizes desktop document research and editing.
ProjectLanguageLicenseStarsStatus
SurfSensePythonApache-2.016.6kActive
local-deep-researchPythonMIT9.2kActive

Considerations

  • Privacy depends in part on model choice. SurfSense keeps its index on disk, but a remote model may receive document content used in a request.
  • Some capabilities have additional requirements. Office-to-PDF conversion needs LibreOffice enabled in Settings, and image creation requires an image model; neither is included or selected by default.
  • OCR support is described for Latin, Chinese, and Japanese scripts. Check suitability if your scanned-document workflow relies on other scripts.
  • The README notes that support for formats beyond PDF, Word, and Excel is still improving.
  • SurfSense is actively developed, with regular commits and releases, but its feature set and format support may continue to change.

Found this useful?

Share it with someone who would like SurfSense.

Comparisons