GitHub listed only ten projects this Thursday morning, and six of them were featured here in the past week. Today's four picks are all first appearances: OpenAI's Codex Security scanner, which shipped version 0.10.0 overnight, Google Chrome's Modern Web Guidance skills for coding agents, shadcn's improve skill that plans with an expensive model and executes with a cheap one, and The Algorithms' long-running Java collection. 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.
Pick 1 · Security / AI code reviewGitHub rank #9 of 10
What it does. Codex Security is OpenAI's CLI and TypeScript SDK for finding, validating and patching vulnerabilities with Codex models. It scans a repository, selected paths or a Git diff, saves a report with findings and a threat model, drafts SECURITY.md policies, exports SARIF, JSON and CSV, and ships a GitHub Action for scheduled scans.
Caveat. Scans are paid model calls sent to an AI provider: the README's example Action sets an estimated $5 limit per scan and warns in-flight requests can exceed it, and 0.10.0 carries breaking changes to paths and MCP setup. It needs a login or API key, some protected findings require OpenAI's Trusted Access for Cyber, and it is for code you are authorised to test. We did not run it.
Attention, not endorsement
GitHub showed
“52 stars today” as displayed for GitHub’s daily window; not an audited rolling 24-hour count
What it does. This is the source repository for Modern Web Guidance, a skill and CLI that feed coding agents current web-platform patterns: View Transitions, container queries, the Popover API, scheduler.yield and some 250 features in all. The agent searches the guides locally and pulls one targeted guide, with fallbacks and Baseline browser-support data, into its context.
Caveat. The eval gains are the maintainers' own figures on their own test apps, not an independent benchmark. The CLI sends anonymous usage telemetry, including agent search queries, to Google unless DISABLE_TELEMETRY=1 is set, and the guide text is CC-BY 4.0 while the code is Apache-2.0. Install it from the published GoogleChrome/modern-web-guidance package, not this source repo. We did not install it.
Attention, not endorsement
GitHub showed
“159 stars today” as displayed for GitHub’s daily window; not an audited rolling 24-hour count
What it does. improve is an agent skill that audits a codebase and writes implementation plans instead of code. A /improve run maps the repo, fans out subagents across nine categories from correctness to direction, re-checks each cited file and line, and writes one self-contained plan per chosen finding into plans/, with verification commands and stop conditions.
Caveat. A full audit with parallel subagents on your most capable model can cost real tokens, so the README offers a cheaper quick pass. Its read-only and no-secrets rules are instructions to the agent, not enforced sandboxing. The repository was last pushed on 12 September 2026 and has no releases, so it is a prompt pack to read before installing. We did not install it.
Attention, not endorsement
GitHub showed
“361 stars today” as displayed for GitHub’s daily window; not an audited rolling 24-hour count
What it does. The Algorithms - Java is a community collection of classic algorithms and data structures written in Java, from sorting and graphs to conversions and ciphers, indexed in a DIRECTORY.md file and built with Maven, with checkstyle, PMD and SpotBugs configured in the repository.
Caveat. The README says the code is for learning, and may be less efficient than the Java standard library; use library or well-tested implementations in production. Quality varies across many contributors, and Trending is no sign of anything new here. We did not run it.
Attention, not endorsement
GitHub showed
“51 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 (): 4 picked today, 6 covered in an earlier digest, 0 not covered. For projects covered before, growth compares the total stars at our first pick with today’s displayed total.
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 only 10 rows at capture, the shortest list of the past week (9 to 18 rows from 5 to 11 November). The capture parsed exactly ten listing rows and is GitHub's own list, not a truncation by us.
All 10 listing entries were inspected. Exact public GitHub API metadata was checked at about 07:01:21 UTC for every candidate, and recent releases for every entry; README files of the four picks were read at about 07:01:27 UTC. Every checked repository was active and unarchived.
Six entries featured in the past seven days were left out without a new primary-sourced reason: meta/astryx (8 November), homeassistant-ai/ha-mcp, vercel-labs/agent-browser and ZhuLinsen/daily_stock_analysis (10 November), and addyosmani/agent-skills and we-promise/sure (11 November).
meta/astryx shipped v0.7.14 on 11 November; its notes describe component behaviour changes (touch-press menus, anchored-layer animation, segmented-control drags), which we did not judge a material change since the 8 November pick. homeassistant-ai/ha-mcp's new releases since 10 November are dev builds (v9.3.0.dev*), also not judged material.
The preceding seven calendar dates, 5-11 November, were checked for local history; all seven editions exist.
The official listing was captured at 08:01:09 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 and eval 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.