GitHub listed fifteen projects this Tuesday morning, and nine of them were featured here in the past week, so they stay out. Today's five picks are all first appearances: Arcane, a self-hosted Docker manager that just shipped version 3; ffmpeg-skill, which gives coding agents a local video editor; Jiangshan Map, an interactive historical atlas of China; Google Research's PaperOrchestra for drafting research papers with agents; and last30days, the social-search skill behind part of this paper's own research. Curation order reflects practical interest, not GitHub rank. Trending is not a release date or an endorsement.
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.
What it does. A self-hosted web interface for managing Docker: containers, images, volumes, Compose projects and remote hosts from one dashboard, written in Go.
Caveat. Version 3 is a breaking upgrade: configuration variables gain an ARCANE_ prefix, legacy protocols, image tags and CLI flags are removed, and the polling agent is gone; the 2.x line lives on a separate branch. Like any Docker manager it needs the Docker daemon, which amounts to root access on the host.
Attention, not endorsement
GitHub showed
“46 stars today” as displayed for GitHub’s daily window; not an audited rolling 24-hour count
What it does. A local video editor for coding agents: 42 FFmpeg tools for cutting, captions, reframing, loudness and exports, packaged as an Agent Skill and an MCP server for Claude Code, Cursor and Codex.
Caveat. It does not ship FFmpeg, so it inherits the bugs and CVEs of whatever build you have installed; its own doctor command only warns. The safety design is the maintainer's description, not something we audited, and the main install path runs an npx package.
Attention, not endorsement
GitHub showed
“367 stars today” as displayed for GitHub’s daily window; not an audited rolling 24-hour count
What it does. An interactive historical atlas of China from 403 BCE to the present: drag a timeline to watch territories change across 18 periods, and click cities to see their names, events and poems through the ages.
Caveat. The README says historical borders are schematic approximations, not surveyed boundaries, and the early first-level divisions have no recorded source. The prefecture layer derives from CHGIS and may not be redistributed or used commercially.
Attention, not endorsement
GitHub showed
“288 stars today” as displayed for GitHub’s daily window; not an audited rolling 24-hour count
What it does. A multi-agent framework from Google Cloud AI Research that turns research notes and experiment logs into a LaTeX paper draft, with agents for the outline, literature review, section writing, refinement and plots.
Caveat. The README calls it not an officially supported Google product, and the dataset used in the paper is not included yet. Running it needs paid model API keys plus a Semantic Scholar key, and a generated draft still needs authors to check every citation and claim.
Attention, not endorsement
GitHub showed
“313 stars today” as displayed for GitHub’s daily window; not an audited rolling 24-hour count
What it does. An agent skill that researches a topic across Reddit, Hacker News, GitHub, YouTube, prediction markets and the web, scores what people engaged with over the last 30 days, and writes a grounded brief.
Caveat. The wider sources rely on your own API keys or browser sessions, which gives an agent access to those accounts. Upvotes and market odds measure attention, not truth, and Polymarket is restricted in some countries, including France.
Attention, not endorsement
GitHub showed
“135 stars today” as displayed for GitHub’s daily window; not an audited rolling 24-hour count
Every repository GitHub listed, in GitHub’s order, with the figures it displayed when the page was checked (): 5 picked today, 9 covered in an earlier digest, 1 not covered. For projects covered before, growth compares the total stars at our first pick with today’s displayed total.
TypeScript · 696 stars today · 12,769 stars in total
The workspace to stay in the loop and on top of agents - shape, decide, plan, annotate, review code, share with team, and send all feedback to agents with one click.
Covered before — first picked 14 Nov 2026 at 10,567 stars; now 12,769: +2,202 stars (+20.8%) in 3 days, about 734 a day.
TypeScript · 1,828 stars today · 28,245 stars in total
Build your own AI office with Univer Office SDK. Bring editing, calculation, collaboration, and file processing into your product for people and AI agents. Built on an extensible open-source core.
Covered before — first picked 15 Nov 2026 at 25,093 stars; now 28,245: +3,152 stars (+12.6%) in 2 days, about 1,576 a day.
Claude Code skills for equity investors and traders — market analysis, technical charting, economic calendars, screeners, and trading strategy development.
Covered before — first picked 14 Nov 2026 at 4,309 stars; now 6,540: +2,231 stars (+51.8%) in 3 days, about 744 a day.
A Claude Code skill bundle for bug hunting and external red-team work - 82 skills, 15 slash commands, 681 disclosed-report patterns curated across 24 core vulnerability classes, plus enterprise identity + infrastructure attack matrices.
Covered before — first picked 16 Nov 2026 at 6,748 stars; now 7,738: +990 stars (+14.7%) in 1 day.
TypeScript · 529 stars today · 9,831 stars in total
A library of Agent Skills designed to work with the Stitch MCP server. Each skill follows the Agent Skills open standard, for compatibility with coding agents such as Antigravity, Gemini CLI, Claude Code, Cursor.
Covered before — first picked 16 Nov 2026 at 9,464 stars; now 9,831: +367 stars (+3.9%) in 1 day.
Multi-tier framework for evaluating AI agent skills and plugins with quality gates, semantic overlap detection, synthetic evaluation dataset generation, and live agent evaluation that measures how skills affect agent behavior.
Covered before — first picked 16 Nov 2026 at 1,964 stars; now 2,982: +1,018 stars (+51.8%) in 1 day.
Local video editor for coding agents: 42 FFmpeg tools with a machine-readable contract and an MCP server, for Claude Code, Cursor and Codex. Cut, captions, reframe, loudness, platform exports. No cloud, no API keys.
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.
GitHub's daily listing held 15 rows at capture. The capture parsed exactly fifteen listing rows and is GitHub's own list, not a truncation by us.
All 15 listing entries were inspected. Exact public GitHub API metadata of the six not-yet-featured entries was read between 07:01:00 and 07:01:04 UTC, their README files between 07:01:11 and 07:01:15 UTC, and recent releases before 07:01:51 UTC. Every checked repository was active and unarchived.
Nine entries featured in the past seven days were left out: ag-ui-protocol/ag-ui and dream-num/univer (15 November), backnotprop/plannotator, tradermonty/claude-trading-skills, alibaba/open-code-review and ChromeDevTools/chrome-devtools-mcp (14 November), and elementalsouls/Claude-BugHunter, google-labs-code/stitch-skills and NVIDIA/SkillEvaluator (16 November). Their releases since those picks are patch builds or dated routine releases, or none at all, which we did not judge a material change.
sveltejs/svelte was inspected and not picked: it is a mature, widely known framework whose latest release, [email protected], dates from 26 October, so the listing gave no new reason to feature it today.
Disclosure: The AI Gazette's twice-daily research step uses an earlier version of the last30days engine (mvanhorn/last30days-skill). It was picked on its own merits, and its card states its limits.
Jiangshan Map is GPL-3.0 for its code and own data, but its prefecture-level boundary layer derives from CHGIS and stays under CHGIS terms (academic, non-commercial use, no redistribution), as its README states. The license field on its card shows GitHub's detection only.
The preceding seven calendar dates, 10-16 November, were checked for local history; all seven editions exist.
The official listing was captured at 08:00:49 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 descriptions, adoption and performance figures come from each project's own README and release notes and are the maintainers' claims. No candidate code, installer, CLI, plugin, skill or model was installed, downloaded or run, and none of this is a security audit or endorsement.