#4848th of 49 in Desktop AI apps

minecontext

Desktop app that screenshots your screen and generates summaries and to-dos

Stars
5.5k
License
Apache-2.0
Last commit
May 2026
Last release
Jan 2026
Language
Python
Stars, last 29 weeks+365
21 Apr 2026: 5,182 stars28 Apr 2026: 5,206 stars5 May 2026: 5,215 stars12 May 2026: 5,233 stars19 May 2026: 5,246 stars26 May 2026: 5,261 stars2 Jun 2026: 5,284 stars9 Jun 2026: 5,312 stars16 Jun 2026: 5,326 stars23 Jun 2026: 5,335 stars30 Jun 2026: 5,349 stars7 Jul 2026: 5,360 stars14 Jul 2026: 5,369 stars21 Jul 2026: 5,387 stars28 Jul 2026: 5,401 stars4 Aug 2026: 5,411 stars11 Aug 2026: 5,419 stars18 Aug 2026: 5,433 stars25 Aug 2026: 5,461 stars1 Sept 2026: 5,469 stars8 Sept 2026: 5,476 stars15 Sept 2026: 5,489 stars22 Sept 2026: 5,498 stars29 Sept 2026: 5,510 stars6 Oct 2026: 5,515 stars13 Oct 2026: 5,517 stars20 Oct 2026: 5,525 stars27 Oct 2026: 5,533 stars3 Nov 2026: 5,539 stars10 Nov 2026: 5,547 stars

5,182 on 21 Apr 20265,547 on 10 Nov 2026

From GitHub's daily star counts.

Checked on GitHub.

Overview

MineContext is an Electron desktop app with a Python FastAPI backend that captures periodic screenshots, interprets them with a vision-language model, and stores the results in SQLite and ChromaDB. It generates daily and weekly summaries, to-dos and activity records in the background, and offers chat over the collected context. Models come from Doubao, OpenAI, or any OpenAI-compatible endpoint, including local ones such as LM Studio.

Who it is for: Knowledge workers wanting automatic screen-based activity summaries and to-dos

Strengths

  • Data stored locally by default; works with any OpenAI-compatible local model endpoint
  • Installers for macOS (.dmg) and Windows (.exe); macOS builds are Apple-notarized
  • Backend debug page shows token usage and lets you edit task intervals and system prompts
  • Backend runs standalone via `uv run opencontext start` with a YAML config

Weaknesses

  • Only screenshots and note editing are marked done; file upload, link upload, meetings are planned
  • Needs a vision-language model and an embedding model; recommended setup uses Doubao cloud API
  • Screenshots are sent to the configured model provider unless a local model is used
  • First run installs the backend environment, taking about two minutes

What it needs

  • Needs SQLite, ChromaDB, uv, Doubao, OpenAI or OpenAI-compatible API
  • Models: Doubao, OpenAI, OpenAI-compatible local models (LM Studio)
  • port 1733

Quick answers

Is minecontext open source?

Yes. Its code is on GitHub under the Apache-2.0 license.

Also in Desktop AI apps

See all 49
Also in Desktop AI apps
RankProjectScore
1cherry-studioDesktop client for chatting with many LLM providers, with agents and knowledge bases53.4k stars+1,124 (+2.1%) in 30 daysAGPL-3.092 out of 100
2dsh-desktopNative Windows and macOS desktop client for DeepSeek Harness, built on plugins35.7k stars+5,734 (+19%) in 30 daysMIT90 out of 100
3screenpipeRecords screen and audio locally, makes it searchable for you and your agents22.3k stars+481 (+2.2%) in 30 dayscustom license90 out of 100
#4prompt-optimizerPrompt rewriting and testing tool for web, desktop, Chrome and MCP Live demo ↗ (opens in a new tab)41.2k stars+3,703 (+9.9%) in 30 dayscustom license86 out of 100
#5aionuiDesktop and WebUI app that runs a built-in agent plus external CLI agents34.2k stars+923 (+2.8%) in 30 daysApache-2.086 out of 100