GitHub Trending · Daily digest · Thursday 5 November 2026luongnv.com/news/trending/reports/2026-11-05.html

GitHub Trending — Thursday 5 November 2026

Source: GitHub Trending, daily (https://github.com/trending?since=daily), checked . GitHub listed 16 repositories; this digest picks 7.

GitHub listed sixteen projects this Thursday morning. NVIDIA's Auto Ontology tops both the list and our picks: a semantic layer that makes data agents show their work. Spec Kit and PR-Agent cover the coding workflow from spec to review, agent-desktop returns because its new release adds an MCP server, and kimi-k3-in-c runs a 2.78-trillion-parameter model from disk on one CPU, slowly. Disc Image Studio is the day's pure delight, and Horizon builds a personal AI news briefing. Curation order reflects practical interest, not GitHub rank. Trending is not a release date or an endorsement.

Curation order, not GitHub rank · Issued · The full list of 16 · Following our picks

The Picks

Picks are listed in curation order, not GitHub’s rank; each card shows the rank GitHub displayed when the page was checked. Stars measure attention, not security, quality or fitness for use, and no project here was installed or run. “Stars today” is GitHub’s own displayed figure for its daily window, not an audited rolling 24-hour count.

Pick 1 · AI agents / data analytics GitHub rank #1 of 16

NVIDIA/auto-ontology

Python · License: Apache-2.0

What it does. An open library from NVIDIA that builds a semantic layer over company databases so AI agents can answer data questions reliably. It generates an ontology (business terms, metrics, relationships and lineage) from existing metadata, reconciles conflicting definitions across systems, turns natural-language questions into SQL, and serves that context to agents over MCP, a CLI and an API. It ships a web UI, Postgres with pgvector, and Docker Compose or Helm deployment.

Caveat. The default setup depends on NVIDIA NIM model endpoints and an NVIDIA API key (hosted or self-hosted), plus Docker, Postgres and optionally HashiCorp Vault for database credentials. An agent connected to databases runs queries with those permissions, so use read-only accounts. The README says the project does not accept external contributions, and it has no tagged release yet. Benchmark results live in a separate evaluation repository. We did not run it.

Attention, not endorsement

GitHub showed
“2,050 stars today” as displayed for GitHub’s daily window; not an audited rolling 24-hour count
Total stars
5,750 at the metadata check

Primary evidence

  1. https://github.com/NVIDIA/auto-ontology
  2. https://api.github.com/repos/NVIDIA/auto-ontology
  3. https://github.com/NVIDIA/auto-ontology/blob/main/LICENSE

Metadata checked . README read from https://api.github.com/repos/NVIDIA/auto-ontology/readme, checked .

Pick 2 · Developer tools / AI coding workflow GitHub rank #7 of 16

github/spec-kit

Python · License: MIT

What it does. GitHub's toolkit that gives AI coding agents structured processes. A Specify CLI sets up a project, and agent commands carry a feature from a written specification through plan, tasks and implementation. Bundled extensions add bug fixing (assessed cause, scoped fix, recorded verification) and idea assessment (go, clarify or stop), while presets, workflows and catalogs let teams adapt or share their own process. It works with Copilot and many other coding agents.

Caveat. It adds process and files to every feature, which can be overkill for small changes. Community catalog components are third-party code, and the README itself says listings are not audits, so review their source before installing. Frequent releases include breaking changes such as v1.3.0's removal, so pin a version. Results still depend on the coding agent you connect. We did not install it.

Attention, not endorsement

GitHub showed
“808 stars today” as displayed for GitHub’s daily window; not an audited rolling 24-hour count
Total stars
147,489 at the metadata check

Primary evidence

  1. https://github.com/github/spec-kit
  2. https://api.github.com/repos/github/spec-kit
  3. https://github.com/github/spec-kit/releases/tag/v1.3.0
  4. https://github.github.io/spec-kit/

Metadata checked . README read from https://api.github.com/repos/github/spec-kit/readme, checked .

Pick 3 · Developer tools / code review GitHub rank #8 of 16

The-PR-Agent/pr-agent

Python · License: MIT

What it does. An open-source AI pull-request reviewer that runs as a GitHub Action, a webhook service, a Docker image or a CLI. Commands such as /describe, /review, /improve and /ask post a PR description, review comments, suggested changes or answers. It supports GitHub, GitLab, Bitbucket, Azure DevOps and Gitea, and any model reachable through LiteLLM, including local models via Ollama.

Caveat. Review quality and cost depend on the model you pay for, and your diffs are sent to that provider. The /help_docs tool has been disabled since v0.36.1 pending a credential-exposure fix, and v0.49.0 changed several defaults (GitHub polling allowlist, Jira project scope, default model), so read the upgrade notes before updating. It is distinct from Qodo's hosted product. We did not run it.

Attention, not endorsement

GitHub showed
“280 stars today” as displayed for GitHub’s daily window; not an audited rolling 24-hour count
Total stars
14,903 at the metadata check

Primary evidence

  1. https://github.com/The-PR-Agent/pr-agent
  2. https://api.github.com/repos/The-PR-Agent/pr-agent
  3. https://github.com/The-PR-Agent/pr-agent/releases/tag/v0.49.0
  4. https://github.com/the-pr-agent/pr-agent/issues/2445

Metadata checked . README read from https://api.github.com/repos/The-PR-Agent/pr-agent/readme, checked .

Pick 4 · AI agents / computer use GitHub rank #6 of 16

vercel-labs/agent-desktop

Rust · License: Apache-2.0

What it does. A command-line tool that lets AI agents operate desktop apps on macOS, Windows and Linux through the operating system's accessibility tree. An agent takes a JSON snapshot with a reference for each element, acts on a reference, then snapshots again to check the result, with no screenshot pixel matching.

Caveat. An easier hookup also makes it easier to hand an agent control of every open app your account can reach, so keep tasks narrow and treat on-screen text as untrusted input. The README warns that snapshots, clipboard reads and screenshots can contain private desktop content. The project is barely a week past v1.0.0, and the MCP server is one day old. We did not install it.

Attention, not endorsement

GitHub showed
“2,470 stars today” as displayed for GitHub’s daily window; not an audited rolling 24-hour count
Total stars
15,729 at the metadata check

Primary evidence

  1. https://github.com/vercel-labs/agent-desktop
  2. https://api.github.com/repos/vercel-labs/agent-desktop
  3. https://github.com/vercel-labs/agent-desktop/releases/tag/v1.2.0

Metadata checked . README read from https://api.github.com/repos/vercel-labs/agent-desktop/readme, checked .

Pick 5 · AI inference / systems programming GitHub rank #11 of 16

FareedKhan-dev/kimi-k3-in-c

C · License: Apache-2.0

What it does. A small, dependency-free C99 inference engine for Moonshot AI's Kimi K3, a 2.78-trillion-parameter mixture-of-experts model. Instead of holding the model in memory, it streams the 1.56 TB checkpoint from disk on each step, so it runs on one CPU with no GPU, BLAS or framework; more RAM only makes it faster. It builds on Linux, macOS and Windows, and its README is a long walkthrough of how each part works.

Caveat. It is very slow by design: the README's own figures are roughly 20 to 33 seconds per token below 128 GB of RAM and 5.6 seconds above, measured on a fast NVMe machine. Running it needs about 1.7 TB of free disk, a Hugging Face token to download the weights, and acceptance of Moonshot AI's model license; the Apache-2.0 license covers only this code. All figures are the author's. We downloaded and ran nothing.

Attention, not endorsement

GitHub showed
“351 stars today” as displayed for GitHub’s daily window; not an audited rolling 24-hour count
Total stars
10,507 at the metadata check

Primary evidence

  1. https://github.com/FareedKhan-dev/kimi-k3-in-c
  2. https://api.github.com/repos/FareedKhan-dev/kimi-k3-in-c
  3. https://github.com/FareedKhan-dev/kimi-k3-in-c/releases/tag/v1.1.0
  4. https://huggingface.co/moonshotai/Kimi-K3

Metadata checked . README read from https://api.github.com/repos/FareedKhan-dev/kimi-k3-in-c/readme, checked .

Pick 6 · Other / hardware hobby GitHub rank #12 of 16

JiaFeiMiao-K-Cat/DiscImageStudio

C# · License: GPL-3.0

What it does. A tool that burns a visible picture onto an ordinary optical disc. It converts an image into data laid out to match the physical tracks of a CD, DVD or single-layer BD-R, so a normal burner writes a pattern you can see by eye on the recording surface. It has a layer-based designer, dithering and tone controls, disc-brand presets, photo-based calibration of track geometry, and exports ISO, RAW or WAV+CUE images or burns directly.

Caveat. The full desktop app is Windows-only; Linux and macOS get an experimental command-line version. Direct streaming burn is experimental, the README says to use only blank, expendable test discs, and the Windows release EXE is unsigned. The latest release notes say a readback problem (issue 90) is not yet confirmed fixed. Results depend on media and drive, and the sample photos are the author's. We did not run it.

Attention, not endorsement

GitHub showed
“1,778 stars today” as displayed for GitHub’s daily window; not an audited rolling 24-hour count
Total stars
6,820 at the metadata check

Primary evidence

  1. https://github.com/JiaFeiMiao-K-Cat/DiscImageStudio
  2. https://api.github.com/repos/JiaFeiMiao-K-Cat/DiscImageStudio
  3. https://github.com/JiaFeiMiao-K-Cat/DiscImageStudio/releases/tag/v1.4.1
  4. https://github.com/JiaFeiMiao-K-Cat/DiscImageStudio/blob/main/README.en.md

Metadata checked . README read from https://api.github.com/repos/JiaFeiMiao-K-Cat/DiscImageStudio/readme, checked .

Pick 7 · Productivity / AI news GitHub rank #15 of 16

Thysrael/Horizon

Python · License: MIT

What it does. A self-hosted, AI-powered news briefing generator. It collects items from RSS, Hacker News, Reddit, Telegram channels, X, GitHub and financial news, scores and deduplicates them with a language model according to per-topic profiles, adds web-researched background and community discussion, and publishes a daily briefing in English and Chinese as Markdown, a GitHub Pages site, email, webhooks or WeChat.

Caveat. Every run spends model API calls, and cost grows with the number of sources and the enrichment you enable. The X source goes through Apify, a paid third-party scraping service, and scraping can conflict with a site's terms. Model-written summaries can be wrong, so keep the links to the originals. There is no tagged release. We did not run it.

Attention, not endorsement

GitHub showed
“121 stars today” as displayed for GitHub’s daily window; not an audited rolling 24-hour count
Total stars
12,703 at the metadata check

Primary evidence

  1. https://github.com/Thysrael/Horizon
  2. https://api.github.com/repos/Thysrael/Horizon
  3. https://www.horizon1123.top
  4. https://thysrael.github.io/Horizon/

Metadata checked . README read from https://api.github.com/repos/Thysrael/Horizon/readme, checked .

The Full List

Every repository GitHub listed, in GitHub’s order, with the figures it displayed when the page was checked (): 7 picked today, 5 covered in an earlier digest, 4 not covered. For projects covered before, growth compares the total stars at our first pick with today’s displayed total.

  1. #1 NVIDIA/auto-ontology

    Python · 2,050 stars today · 5,744 stars in total

    Auto Ontology is a Python framework for building semantic, graph-based reasoning layers over data. It enables deterministic query generation, better context retrieval, and more explainable AI for LLM and agent workflows.

    Picked today — Pick 1 above.

  2. #2 max-sixty/worktrunk

    Rust · 531 stars today · 11,188 stars in total

    Worktrunk is a CLI for Git worktree management, designed for parallel AI agent workflows

    Covered before — first picked 3 Nov 2026 at 10,156 stars; now 11,188: +1,032 stars (+10.2%) in 2 days, about 516 a day.

  3. #3 mattpocock/skills

    Shell · 1,379 stars today · 313,731 stars in total

    Skills for Real Engineers. Straight from my .agents directory.

    Covered before — first picked 31 Oct 2026 at 307,678 stars; now 313,731: +6,053 stars (+2.0%) in 5 days, about 1,211 a day.

  4. #4 floci-io/floci

    Java · 1,090 stars today · 29,818 stars in total

    Light, fluffy, and always free - The AWS Local Emulator alternative

    Covered before — first picked 4 Nov 2026 at 29,055 stars; now 29,818: +763 stars (+2.6%) in 1 day.

  5. #5 thedotmack/claude-mem

    TypeScript · 1,172 stars today · 113,086 stars in total

    Persistent Context Across Sessions for Every Agent – Captures everything your agent does during sessions, compresses it with AI, and injects relevant context back into future sessions. Works with Claude Code, OpenClaw, Codex, Gemini, Hermes, Copilot, OpenCode + More

    Covered before — first picked 30 Oct 2026 at 105,064 stars; now 113,086: +8,022 stars (+7.6%) in 6 days, about 1,337 a day. Also picked 2 Nov 2026.

  6. #6 vercel-labs/agent-desktop

    Rust · 2,470 stars today · 15,727 stars in total

    Computer use CLI for AI agents

    Picked today — Pick 4 above; first covered 3 Nov 2026 at 10,821 stars; now 15,727: +4,906 stars (+45.3%) in 2 days, about 2,453 a day.

  7. #7 github/spec-kit

    Python · 808 stars today · 147,489 stars in total

    Toolkit to help you get started with SDD or any other process!

    Picked today — Pick 2 above.

  8. #8 The-PR-Agent/pr-agent

    Python · 280 stars today · 14,903 stars in total

    PR Agent: The Original Open-Source PR Reviewer. This project is not the Qodo free tier.

    Picked today — Pick 3 above.

  9. #9 harry0703/MoneyPrinterTurbo

    Python · 617 stars today · 139,278 stars in total

    利用 AI 大模型和自动化工作流,根据主题或关键词一键生成高清短视频。Generate HD short videos from a topic or keyword with an automated AI workflow.

    Not covered yet.

  10. #10 garrytan/gstack

    TypeScript · 290 stars today · 139,499 stars in total

    Use Garry Tan's exact Claude Code setup: 23 opinionated tools that serve as CEO, Designer, Eng Manager, Release Manager, Doc Engineer, and QA

    Not covered yet.

  11. #11 FareedKhan-dev/kimi-k3-in-c

    C · 351 stars today · 10,507 stars in total

    A 2.78-trillion-parameter Kimi K3 running inference on a single CPU in 8.24 GB of RAM. Portable C99: no BLAS, no framework, no GPU.

    Picked today — Pick 5 above.

  12. #12 JiaFeiMiao-K-Cat/DiscImageStudio

    C# · 1,778 stars today · 6,820 stars in total

    Picked today — Pick 6 above.

  13. #13 MichaelCade/90DaysOfDevOps

    Shell · 97 stars today · 30,605 stars in total

    This repository started out as a learning in public project for myself and has now become a structured learning map for many in the community. We have 3 years under our belt covering all things DevOps, including Principles, Processes, Tooling and Use Cases surrounding this vast topic.

    Not covered yet.

  14. #14 D4Vinci/Scrapling

    Python · 2,005 stars today · 96,970 stars in total

    The BEST Web Scraping framework that handles everything from a single request to a full-scale crawl! Don't be shy, join here: https://discord.gg/EMgGbDceNQ and follow here for daily tips and tricks: https://x.com/Scrapling_dev

    Covered before — first picked 3 Nov 2026 at 93,362 stars; now 96,970: +3,608 stars (+3.9%) in 2 days, about 1,804 a day.

  15. #15 Thysrael/Horizon

    Python · 121 stars today · 12,703 stars in total

    Your own AI-powered news radar. Generates daily briefings in English & Chinese. | 用 AI 构建你专属的新闻雷达

    Picked today — Pick 7 above.

  16. #16 scenario-labs/skills

    Python · 224 stars today · 2,608 stars in total

    Get production-ready images, video, audio, and 3D from any AI agent: skills that pick the right model, price before spending, and keep characters and brands consistent through the Scenario MCP, plus expert teams that drive Blender, Maya, ZBrush, Unreal, and Unity.

    Not covered yet.

Following Our Picks

Every project picked in an earlier digest, biggest growth first. “Then” is the total at our first pick’s metadata check; “now” is GitHub’s API star count checked . Growth measures attention, not quality or safety.

ProjectFirst pickedStars thenStars nowChangePer daySince last check
morluto/rea15 Oct 2026250,334431,956+181,622 (+72.6%)8,649+3,814 since 4 Nov 2026
debpalash/VoiceStudio (picked 3 times)15 Oct 202667,106101,250+34,144 (+50.9%)1,626+926 since 4 Nov 2026
cloudflare/security-audit-skill21 Oct 202634,63859,821+25,183 (+72.7%)1,679+1,437 since 4 Nov 2026
MadsLorentzen/ai-job-search (picked 2 times)17 Oct 202648,78969,670+20,881 (+42.8%)1,099+233 since 4 Nov 2026
stablyai/orca18 Oct 202696,443115,816+19,373 (+20.1%)1,076+1,029 since 4 Nov 2026
affaan-m/ECC16 Oct 2026280,607297,412+16,805 (+6.0%)840+695 since 4 Nov 2026
pbakaus/impeccable25 Oct 202686,664103,123+16,459 (+19.0%)1,496+1,329 since 4 Nov 2026
Nutlope/hallmark25 Oct 202632,49848,172+15,674 (+48.2%)1,425+936 since 4 Nov 2026
DietrichGebert/ponytail21 Oct 2026170,170184,723+14,553 (+8.6%)970+777 since 4 Nov 2026
pingdotgg/t3code31 Oct 202633,06546,990+13,925 (+42.1%)2,785+2,063 since 4 Nov 2026
hmjz100/LinkSwift17 Oct 202623,96137,866+13,905 (+58.0%)732+156 since 4 Nov 2026
xerj-org/xerj16 Oct 20265,93619,121+13,185 (+222.1%)659+145 since 4 Nov 2026
Graphify-Labs/graphify21 Oct 2026130,551142,575+12,024 (+9.2%)802+584 since 4 Nov 2026
obra/superpowers (picked 2 times)17 Oct 2026300,784312,585+11,801 (+3.9%)621+577 since 4 Nov 2026
NationalSecurityAgency/ghidra15 Oct 202687,83797,924+10,087 (+11.5%)480+304 since 4 Nov 2026
Renhuai123/ziwei-doushu24 Oct 20267,01815,929+8,911 (+127.0%)743+599 since 4 Nov 2026
HKUDS/DeepTutor28 Oct 202646,60355,138+8,535 (+18.3%)1,067+303 since 4 Nov 2026
anomalyco/opencode18 Oct 2026216,421224,654+8,233 (+3.8%)457+225 since 4 Nov 2026
thedotmack/claude-mem (picked 2 times) (on today’s list, #5)30 Oct 2026105,064113,087+8,023 (+7.6%)1,337+881 since 4 Nov 2026
anthropics/claude-plugins-official16 Oct 202639,40746,164+6,757 (+17.1%)338+137 since 4 Nov 2026
Comfy-Org/ComfyUI (picked 2 times)18 Oct 2026138,516145,096+6,580 (+4.8%)366+178 since 4 Nov 2026
mattpocock/skills (on today’s list, #3)31 Oct 2026307,678313,737+6,059 (+2.0%)1,212+1,936 since 4 Nov 2026
Donchitos/Claude-Code-Game-Studios30 Oct 202629,60135,557+5,956 (+20.1%)993+639 since 4 Nov 2026
TauricResearch/TradingAgents26 Oct 2026114,951120,692+5,741 (+5.0%)574+239 since 4 Nov 2026
JetBrains/thinkrail26 Oct 20262,0817,671+5,590 (+268.6%)559+193 since 4 Nov 2026
HKUDS/CLI-Anything15 Oct 202652,71458,273+5,559 (+10.5%)265+74 since 4 Nov 2026
AgriciDaniel/claude-seo26 Oct 202621,32726,761+5,434 (+25.5%)543+345 since 4 Nov 2026
K-Dense-AI/scientific-agent-skills16 Oct 202650,75156,138+5,387 (+10.6%)269+116 since 4 Nov 2026
LibreSpark/LibreTV16 Oct 202618,26423,570+5,306 (+29.1%)265+86 since 4 Nov 2026
ruvnet/ruflo (picked 2 times)17 Oct 202675,72080,911+5,191 (+6.9%)273+91 since 4 Nov 2026
makecindy/cindy25 Oct 20264,2699,304+5,035 (+117.9%)458+102 since 4 Nov 2026
shiyu-coder/Kronos20 Oct 202641,84046,870+5,030 (+12.0%)314+244 since 4 Nov 2026
vercel-labs/agent-desktop (picked 2 times) (on today’s list, #6)3 Nov 202610,82115,733+4,912 (+45.4%)2,456+2,309 since 4 Nov 2026
Panniantong/Agent-Reach27 Oct 2026104,122108,928+4,806 (+4.6%)534+395 since 4 Nov 2026
akitaonrails/ai-memory (picked 2 times)18 Oct 202610,39515,163+4,768 (+45.9%)265+100 since 4 Nov 2026
milind-soni/OpenMausBot30 Oct 20267,70312,293+4,590 (+59.6%)765+497 since 4 Nov 2026
elayadesign/ai-design-skills28 Oct 20269,99114,209+4,218 (+42.2%)527+184 since 4 Nov 2026
mksglu/context-mode15 Oct 202629,34633,509+4,163 (+14.2%)198+98 since 4 Nov 2026
nightly-labs/openbot27 Oct 20261,6975,635+3,938 (+232.1%)438+278 since 4 Nov 2026
anthropics/knowledge-work-plugins21 Oct 202630,88934,770+3,881 (+12.6%)259+122 since 4 Nov 2026
xtool-org/xtool15 Oct 202611,14114,760+3,619 (+32.5%)172+137 since 4 Nov 2026
D4Vinci/Scrapling (on today’s list, #14)3 Nov 202693,36296,975+3,613 (+3.9%)1,806+1,654 since 4 Nov 2026
vxcontrol/pentagi15 Oct 202629,88633,359+3,473 (+11.6%)165+134 since 4 Nov 2026
luongnv89/claude-howto23 Oct 202643,46646,548+3,082 (+7.1%)237+74 since 4 Nov 2026
lightpanda-io/browser31 Oct 202638,62841,326+2,698 (+7.0%)540+139 since 4 Nov 2026
headroomlabs-ai/headroom20 Oct 202676,37078,883+2,513 (+3.3%)157+102 since 4 Nov 2026
pollen-robotics/microduck25 Oct 202611,15413,455+2,301 (+20.6%)209+45 since 4 Nov 2026
gitroomhq/postiz-app27 Oct 202638,67940,932+2,253 (+5.8%)250+88 since 4 Nov 2026
android/skills17 Oct 20268,14910,343+2,194 (+26.9%)115+28 since 4 Nov 2026
NVIDIA/OpenShell24 Oct 202618,93421,111+2,177 (+11.5%)181+50 since 4 Nov 2026
Orkas-AI/Orkas1 Nov 20266,8809,025+2,145 (+31.2%)536+32 since 4 Nov 2026
anthropics/skills31 Oct 2026185,520187,507+1,987 (+1.1%)397+316 since 4 Nov 2026
NVIDIA/SkillSpector27 Oct 202622,46324,410+1,947 (+8.7%)216+154 since 4 Nov 2026
CopilotKit/OpenBot30 Oct 20268,32910,188+1,859 (+22.3%)310+124 since 4 Nov 2026
vllm-project/vllm21 Oct 202694,53296,364+1,832 (+1.9%)122+97 since 4 Nov 2026
anthropics/claude-code20 Oct 2026152,763154,580+1,817 (+1.2%)114+125 since 4 Nov 2026
SnailSploit/Claude-Red24 Oct 20268,62310,370+1,747 (+20.3%)146+42 since 4 Nov 2026
alchaincyf/huashu-design27 Oct 202627,28628,725+1,439 (+5.3%)160+102 since 4 Nov 2026
supabase/supabase22 Oct 2026113,145114,542+1,397 (+1.2%)100+45 since 4 Nov 2026
withmarbleapp/os-taxonomy20 Oct 20266,2827,598+1,316 (+20.9%)82+23 since 4 Nov 2026
ruvnet/RuView29 Oct 2026101,281102,455+1,174 (+1.2%)168+152 since 4 Nov 2026
twentyhq/twenty22 Oct 202659,34860,463+1,115 (+1.9%)80+62 since 4 Nov 2026
JustVugg/colibri3 Nov 202646,76447,868+1,104 (+2.4%)552+455 since 4 Nov 2026
FlashML-org/FreeToken2 Nov 202615,59216,649+1,057 (+6.8%)352+111 since 4 Nov 2026
max-sixty/worktrunk (on today’s list, #2)3 Nov 202610,15611,191+1,035 (+10.2%)518+533 since 4 Nov 2026
coollabsio/coolify26 Oct 202664,67565,652+977 (+1.5%)98+55 since 4 Nov 2026
mlabonne/llm-course18 Oct 202683,95984,912+953 (+1.1%)53+19 since 4 Nov 2026
vorssaint/vorssaint-utils1 Nov 202632,05632,988+932 (+2.9%)233+164 since 4 Nov 2026
beenuar/AiSOC2 Nov 20266,4137,300+887 (+13.8%)296+253 since 4 Nov 2026
hyprwm/Hyprland2 Nov 202639,51040,317+807 (+2.0%)269+72 since 4 Nov 2026
Meloong-Git/PCL23 Oct 20268,7779,569+792 (+9.0%)61+16 since 4 Nov 2026
floci-io/floci (on today’s list, #4)4 Nov 202629,05529,822+767 (+2.6%)767—
cjpais/Handy2 Nov 202636,82837,567+739 (+2.0%)246+165 since 4 Nov 2026
OpenBMB/VoxCPM24 Oct 202640,36241,054+692 (+1.7%)58+42 since 4 Nov 2026
paperclipai/paperclip3 Nov 2026110,580111,197+617 (+0.6%)308+299 since 4 Nov 2026
macro-inc/macro29 Oct 20265,5556,152+597 (+10.7%)85+22 since 4 Nov 2026
fastapi/fastapi23 Oct 2026103,901104,414+513 (+0.5%)39+24 since 4 Nov 2026
google-gemini/gemini-cli25 Oct 2026108,241108,733+492 (+0.5%)45+19 since 4 Nov 2026
php/php-src23 Oct 202641,29241,759+467 (+1.1%)36+0 since 4 Nov 2026
kornia/kornia2 Nov 202611,90912,338+429 (+3.6%)143+20 since 4 Nov 2026
Usagi-org/ai-goofish-monitor4 Nov 202616,79417,218+424 (+2.5%)424—
THU-MAIC/OpenMAIC1 Nov 202642,28442,706+422 (+1.0%)106+76 since 4 Nov 2026
xiaolai/the-craft-of-selfteaching29 Oct 202618,70219,098+396 (+2.1%)57+28 since 4 Nov 2026
expo/skills29 Oct 20263,6774,054+377 (+10.3%)54+13 since 4 Nov 2026
bregman-arie/devops-exercises1 Nov 202686,13586,510+375 (+0.4%)94+31 since 4 Nov 2026
anomalyco/hex4 Nov 20264,7605,119+359 (+7.5%)359—
lightseekorg/tokenspeed16 Oct 20262,5602,865+305 (+11.9%)15+9 since 4 Nov 2026
nodejs/node24 Oct 2026122,952123,239+287 (+0.2%)24+13 since 4 Nov 2026
karpathy/nanochat3 Nov 202660,04260,278+236 (+0.4%)118+60 since 4 Nov 2026
aws/agent-toolkit-for-aws25 Oct 20263,2093,404+195 (+6.1%)18+13 since 4 Nov 2026
spotify/portal-ai-plugins4 Nov 20263,0993,155+56 (+1.8%)56—

Source coverage & colophon

Source

Period
daily, as defined by GitHub
Checked
Repositories listed
16

Coverage notes

  • All 16 listing entries were inspected. Exact public GitHub API metadata was checked for the ten entries never featured before and for vercel-labs/agent-desktop at about 07:01 UTC, recent releases for 13 entries, and README files for eleven; every checked repository was active and unarchived. All seven picks have a licence detected by GitHub's API (Apache-2.0, MIT or GPL-3.0).
  • One repeat: vercel-labs/agent-desktop (3 November), because v1.2.0, published at 02:43 UTC today, adds an MCP server; the change is described on its card.
  • Five entries featured before were left out without a material change: max-sixty/worktrunk and D4Vinci/Scrapling (3 November), floci-io/floci (4 November), thedotmack/claude-mem (30 October and 2 November) and mattpocock/skills (31 October). Worktrunk, Floci and mattpocock/skills have had no release since; claude-mem's v15.9.0 was already noted on 3 November; Scrapling's v0.5.1 (4 November evening) adds agent drag-and-drop and remote browser sessions but was judged an incremental follow-up to the v0.5 covered then.
  • New entries not picked: harry0703/MoneyPrinterTurbo (an established automatic short-video generator whose README leads with model-API sponsor offers), garrytan/gstack (a personal Claude Code skill set close to setups covered before), scenario-labs/skills (agent skills that mostly need a Scenario account) and MichaelCade/90DaysOfDevOps (a DevOps learning path last updated in June, whose licence GitHub's API reports as Other).
  • No security-focused project was on today's list.
  • The preceding seven calendar dates, 29 October-4 November, were checked for local history; all seven editions exist.
  • The official listing was captured at 08:00:42 Europe/Paris. Literal stars-today labels are GitHub's displayed period figures; API total stars and README evidence were checked separately at the times shown per card. They are not an independently audited rolling 24-hour growth measurement.
  • Feature, speed and quality descriptions are the maintainers' own, including kimi-k3-in-c's memory and per-token timings, PR-Agent's single-call claim and Disc Image Studio's sample results. No candidate code, installer, extension, checkpoint or model was installed, downloaded or run, and none of this is a security audit or endorsement.