GitNova

GitHub seismograph — discoveries and growing interest

GitNova Daily discoveries — November 3, 2026 · 22:00 MSK

Three discoveries with different uses, selected from the projects GitNova tracks. This edition was saved on November 3, 2026, 19:00 UTC.

Summaries come from README digests. Magnitude measures growing interest; installation has not been tested. Star counts below refer to the stated UTC day at the time shown.

rerun-build/rerun

AI agents · TypeScript

+323stars on 2026-11-03 · UTC

Platform for hiring AI coworkers: a Foreman plans the team, each agent gets its own memory, skills and computer, connecting to MCP and 200+ integrations. Self-hosted, AGPL-3.0.

Useful for
  • Self-host Rerun with Docker and Supabase on your own server
  • Set up an AI coworker to triage inbox and calendar
  • Connect an MCP server or Composio integration to an agent
Before using
  • License: AGPL-3.0
  • Installation and requirements have not been verified; review the README before running
Explore project

Why now

Created on GitHub
2026-10-29 UTC
First seen by GitNova
2026-11-03 UTC
Magnitude
7.3 / 10 · growing interest

Only 4 complete days of history; a reliable usual pace is not established. On 2026-11-02: +332 stars (UTC)

Maximum over those 4 complete days: +332 on 2026-11-02; the current measurement has not exceeded it (UTC)

Measured: November 3, 2026, 18:16 UTC

Measurement details
  • Created on GitHub: 2026-10-29; first seen by GitNova: 2026-11-03 (UTC)
  • +323 stars measured 2026-11-03, 00:31–17:59 UTC
  • Only 4 complete days of history; a reliable usual pace is not established. On 2026-11-02: +332 stars (UTC)
  • Maximum over those 4 complete days: +332 on 2026-11-02; the current measurement has not exceeded it (UTC)
  • Magnitude 7.3/10

karpathy/nanochat

Language models · Python

+287stars on 2026-11-03 · UTC

A minimal experimental harness for training LLMs on a single GPU node, covering tokenization, pretraining, finetuning, evaluation and inference. It lets you train a GPT-2-capability model for about $48 and chat with it via CLI.

Useful for
  • Train your own GPT-2-capability model on 8XH100 and chat with it via CLI
  • Set model complexity with a single --depth flag and get auto-tuned hyperparameters
  • Reproduce the GPT-2 speedrun and compare training time against the leaderboard
Before using
  • License: MIT
  • Installation and requirements have not been verified; review the README before running
Explore project

Why now

Created on GitHub
2025-10-13 UTC
First seen by GitNova
2026-11-02 UTC
Typical day
+30 stars · median of 28 complete days
Magnitude
7.0 / 10 · growing interest

Already 9.6× a typical complete day; a spike in interest, not a quality score

Maximum over those 28 complete days: +512 on 2026-11-02; the current measurement has not exceeded it (UTC)

Measured: November 3, 2026, 18:16 UTC

Measurement details
  • Created on GitHub: 2025-10-13; first seen by GitNova: 2026-11-02 (UTC)
  • +287 stars measured 2026-11-03, 00:30–17:59 UTC
  • Usually: +30 stars/day — median of 28 complete days 2026-10-06–2026-11-02 (UTC)
  • Already 9.6× a typical complete day; a spike in interest, not a quality score
  • Maximum over those 28 complete days: +512 on 2026-11-02; the current measurement has not exceeded it (UTC)
  • Magnitude 7.0/10

hyprwm/Hyprland

Other · C++

+275stars on 2026-11-03 · UTC

Hyprland is an independent, dynamic tiling Wayland compositor written in C++ with a focus on visual effects and customization. It is not based on wlroots and offers plugins, IPC, and multiple layout modes.

Useful for
  • Replace an existing Wayland compositor with a tiling setup that supports animations and blur
  • Configure per-workspace layouts and window rules via the config file
  • Write a plugin to add custom window management behavior
Before using
  • License: BSD-3-Clause
  • Installation and requirements have not been verified; review the README before running
Explore project

Why now

Created on GitHub
2022-03-16 UTC
First seen by GitNova
2026-10-30 UTC
Typical day
+14.5 stars · median of 28 complete days
Magnitude
6.8 / 10 · growing interest

Already 19.0× a typical complete day; a spike in interest, not a quality score

Maximum over those 28 complete days: +564 on 2026-11-02; the current measurement has not exceeded it (UTC)

Measured: November 3, 2026, 18:16 UTC

Measurement details
  • Created on GitHub: 2022-03-16; first seen by GitNova: 2026-10-30 (UTC)
  • +275 stars measured 2026-11-03, 00:30–17:59 UTC
  • Usually: +14.5 stars/day — median of 28 complete days 2026-10-06–2026-11-02 (UTC)
  • Already 19.0× a typical complete day; a spike in interest, not a quality score
  • Maximum over those 28 complete days: +564 on 2026-11-02; the current measurement has not exceeded it (UTC)
  • Magnitude 6.8/10

Share this edition

Share on XTelegram

Previous editions

Three GitHub discoveries in each edition

What they do, why they are gaining interest, and what to check before using them.

Editions at 10:00, 16:00, 22:00 MSK.

Read daily discoveries in English Project updates · English RSS Telegram · in Russian