Recently updated open-source projects

Construction Plan Review checks permit sets sheet by sheet across disciplines, then coordinates contradictions between sheets and compares revisions. For active jobs it tracks revised sheets, ASIs, RFIs, shop drawings, inspections, and plan-versus-field questions, returning either a structured issue log or a short action list. Regulatory points are marked for verification rather than declared compliant, and the skill is designed for plan checkers and owner representatives.

mastery-loop ingests slides, textbooks, PDFs, notes, or a course folder, builds a map from easier to harder material, and creates active-recall items for a daily review session. A real FSRS scheduler interleaves questions, waits for the learner’s answer, handles errors and Feynman checks, and adapts future reviews. Knowledge stays in Markdown and JSON owned by the user, while the scheduler needs Python’s standard library and readers for any binary formats.

Reconstruct a development timeline from local Claude Code conversations, filtering out tool noise while preserving attempts, setbacks, fixes, and turning points. Vibe Story then turns that timeline into a first-person dev log, technical retrospective, or longer article, with optional date and topic filters. Both the skeleton and finished article stay inside the target project, but the extractor currently supports Claude Code logs only.

atscale-prd-writer helps enterprise B2B data-platform teams turn a problem into a reviewable, buildable product document. It checks the problem statement, testable requirements, success metrics, and acceptance criteria, while keeping migration compatibility, API stability, multi-persona conflicts, and the operator control plane explicit and separating PRDs from design documents.

auditoria-ley-21719 examines a project’s stack, database models, migrations, and forms to identify personal or sensitive data, then checks 17 categories under Chile’s Law 21.719. It produces a prioritized Markdown report with file-and-line references and labels each item as implemented, partial, missing, not applicable, or requiring confirmation when the repository cannot show the answer. The package is a technical compliance aid, not legal advice, especially for high-risk processing.

Foreman wraps another skill in a guided session: it explains the target's scope and inputs, exposes risks, shows progress, and pauses at decisions that could change the result. Users can choose an option or let it proceed. Because it is a prompt-only skill, long conversations may weaken supervision and there is no persistent dashboard or timer.

vault is a desktop organizer for links, Markdown notes, and original-quality images stored in ordinary local folders. Its browser extension sends saved content to the desktop app through a local API, where items can be grouped into folders or separate vaults, searched, renamed, pinned, imported, and exported. It works without an internet connection and does not include built-in synchronization, although users can sync the vault folder themselves with another storage service.

Run axe-core through Playwright against public or session-authenticated pages, map each violation to the relevant WCAG success criteria, and use the application source to propose concrete edit diffs. The skill produces a deterministic Markdown findings report and can optionally fill a user-supplied VPAT 2.5 INT template. It is primarily intended for auditing web apps whose source Claude can read.

Document Cloning Skills turns Markdown into tightly controlled Word documents for tenders, government papers, technical proposals, and white papers. Skeleton mode preserves the reference DOCX’s styles, numbering, headers, footers, theme, and page setup while replacing the body; profile mode uses a preset or an extracted style profile. The renderer also handles cover and contents sections, page breaks, table headers, and separate Chinese and Latin fonts.

Job Trawlers Interview Prep gathers the target role, your background, interview stage, and interviewers, then researches the company and people to build a tailored preparation kit. It maps likely questions and follow-ups to STAR stories, suggests sharp questions to ask back, and runs a one-question-at-a-time mock interview with candid feedback. The downloadable skill is released under CC BY 4.0, so reuse and adaptation require keeping the required attribution.

Instagram Reels Gallery reads a selected Instagram DM group, records which friend shared each Reel, downloads local thumbnails, and builds a static site with filters, date sorting, and links back to Instagram. The result can open from a local file or be deployed to a static host such as Netlify. Authentication uses the account’s sessionid rather than a password, so the credential must be protected and revoked like a password when no longer needed.

Instagram Reels Transcriber converts one Reel or all Reels from a profile into searchable Markdown notes with dates, likes, comments, views, shares, captions, and transcripts. It uses yt-dlp and Groq Whisper with automatic language detection. A single Reel needs no Instagram login, but transcription still needs a Groq API key; profile collection additionally requires the user’s Instagram session cookie. Existing notes can be skipped or have their metadata refreshed.

TikTok Transcriber accepts one TikTok URL or an entire creator profile, turns each video into a clean Markdown note, and records its date, views, likes, comments, reposts, and saves. Audio is transcribed with Groq Whisper using the user’s Groq API key, and the plain Markdown works in Obsidian or elsewhere. It needs no TikTok login, cookies, or browser automation; reruns skip notes that already exist.

infisical-cli-secrets teaches an agent to use Infisical CLI as a secret boundary: authenticate as a user or machine identity, inject secrets into child processes, and verify behavior without exposing values. It also helps decide when a value belongs in gateway environment variables versus an application-owned encrypted workspace store. The method is aimed at development services, CI, and Pawrrtal workflows where the ownership and exposure boundary must remain explicit.

wordpress-page-builder-architecture describes a no-build WordPress visual editor built with PHP and vanilla JavaScript. A single PHP renderer serves both the front end and a same-origin iframe canvas, while a JSON node tree in post metadata holds the page model and one control schema drives sanitization and editing. The guidance also covers cacheable CSS files, REST and nonce setup, widget registration, and keeping editor and front-end output aligned.

webcpu-agent-skill gives coding agents verified references, examples, and templates for building browser-based Canvas2D scenes with EASEL.js 0.5.0. It packages the same guidance for Claude, Codex, OpenCode, Copilot, and other agent environments, with a rule to report unknown APIs rather than invent them. Rendering still requires a browser DOM canvas; Deno can type-check, bundle, and serve the app but does not provide headless Canvas2D by itself.

Turn an assignment prompt into a rubric that contains everything the evaluator needs, audits omissions and contradictions, and keeps score-bearing criteria separate from checklist subcriteria. The skill can also simulate a real correction locally with the Claude CLI and validate the result against backend rules. Python 3.10 is required, and TEST mode needs `claude` on PATH.

github-auto-backup interviews you about repository settings when creating a project, then manages local Git and GitHub backups with conventional commits. It can commit, push, pull, and create branches, but pauses before public pushes, force-pushes, opening or merging pull requests, deleting branches, or changing repository settings. It never commits secrets or silently makes a private repository public. Local-only mode works without GitHub; GitHub operations need an authenticated gh CLI.