Recently updated open-source projects

Use this skill or CLI to turn PDFs, office files, and images into fax-native 1-bit CCITT-G4 PDF or multipage TIFF while protecting small text, signatures, forms, and scans. It can produce comparison sheets and per-page reports so you can inspect readability before transmission. Office inputs require LibreOffice, while optional OCR improves text embedded in images at a substantial processing-time cost.

PDF Email Optimizer converts common Office files when needed, then reduces the resulting PDF toward a requested attachment limit through structural cleanup and controlled image recompression. It keeps the source untouched, reports the chosen strategy and quality trade-offs, and can render before-and-after pages for visual checks. Office inputs require LibreOffice, while unusually tight size targets may require an explicit lossy compression mode.

PDF Bookmarks reads a printed table of contents from scanned or OCR’d books and writes matching PDF bookmarks and page labels. It corrects front-matter offsets between printed and PDF pages, and can fall back to font-based detection or a readable manual list when parsing is uncertain. OCR options include Tesseract, macOS Vision, and Windows OCR, but the Windows backend is still beta and unverified on real Windows hardware.

Hangeul DeAI edits Korean prose for translation-like phrasing, excess commas, repetitive endings, buzzwords, and a mismatch between the writer, document, and audience. It favors targeted edits over wholesale rewriting and checks that numbers, proper names, and claims remain unchanged. The skill is for natural, trustworthy writing, not for bypassing AI detectors or adding artificial mistakes.

Quick Picture Viewer is a lightweight Windows image viewer that opens common formats along with files such as WebP, SVG, DDS, PSD, and CR2. Users can rotate, flip, crop, convert, run slideshows, use picture-in-picture mode, print images, or export them to PDF. Plugin support adds extra effects and tools, while the Windows-only scope makes it primarily an alternative to the default photo viewer.

Draw.io Diagrammer generates flowcharts, SOPs, ERDs, architecture diagrams, UML class diagrams, and sequence diagrams from reusable XML templates. Every render goes through a screenshot, inspect, fix, and re-export loop to catch overflow, cramped spacing, collisions, and bad arrow routing. It requires the draw.io desktop binary; Linux or headless runs also need xvfb.

localize-infographic-pdf uses a hidden searchable text layer to locate visible vector or Type3 text, erases each source run, and redraws the translation in the same position, color, weight, and size. Photos, charts, gradients, icons, and logos are not regenerated. It targets Canva, Figma, or Illustrator one-pagers and follows analyze, extract, compose, and PDF stages with translator and reviewer agents.

Use targeted techniques such as blind-spot scouting, one-question-at-a-time interviews, reference-code inspection, divergent prototypes, planning notes, and post-change quizzes to clarify ambiguous work. For higher-stakes tasks, Cartographer mode keeps an unknowns ledger, prioritizes gaps by regret, and blocks implementation until coverage and ambiguity gates pass. The framework runs as Markdown inside Claude Code, with optional specialist agents.

This repository packages reusable Agent Skills for writing product requirements and reviewing completed implementations with Codex or compatible tools. You can install an individual skill through `npx skills`, GitHub CLI, or the Codex skill installer, either globally or for a specific environment. It also includes copyable `AGENTS.md` and `CLAUDE.md` templates for projects that need consistent agent instructions. The value is a small, focused workflow library rather than a standalone application or general-purpose agent runtime.

recipe-extract-skill extracts a recipe’s title, ingredients, instructions, yield, and times from a URL, using `schema.org/Recipe` JSON-LD as the fast path and asking Claude to fill only missing fields when needed. It cleans the page before rendering a concise result and includes a per-path evaluation suite with cached inputs. Pages without usable structured data require the slower LLM path.

AGX brings tickets, repositories, agent sessions, implementation work, and pull-request review into a local CLI and dashboard with durable checkpoints for resuming later. It supports Claude, Codex, Gemini, and Ollama with human approval gates before irreversible actions; the CLI needs Node 22.16 or newer and at least one provider CLI, while anonymous telemetry is enabled by default but can be disabled.

Build a personal knowledge base on local plain-text files without creating an account. The app combines Markdown, WYSIWYG, split-view, and mind-map editing with full-text search, graph, task, timeline, folder, backlink, and Kanban views. An integrated reader brings RSS, Atom, and podcast subscriptions into the same workspace, while documents can include diagrams, equations, code, and local attachments. Desktop builds are available for Windows, macOS, and Linux, with a web version as well; mobile clients are still listed as future work.

Reel Check downloads and transcribes Instagram, TikTok, or YouTube Shorts locally, extracts the video’s concrete claims, and checks named projects before judging the idea. It also compares the concept with your existing skills, automations, and configuration, then returns an already-have, implement, partial, skip, or interesting-only verdict. It never implements the recommendation automatically.

inglite-china-trekking searches real China treks by region, difficulty, and length, then provides route details, deep guides, and China-specific logistics such as permits, altitude, payments, and seasons. With the traveller’s consent, its booking action sends a real trip enquiry to Inglite’s English-speaking concierge. It provides route data but deliberately no downloadable GPS or GPX tracks; treks are walked with a local guide.

de-ai-gate scans HTML, prose, and authored records for typographic and phrase patterns associated with machine-written copy, then exits clean or failed with a deterministic code. It writes nothing, uses no network or credentials, and can run from Node, Bun, or Python in a pre-ship or CI gate. It reports offending tells without modifying the files.

No Fabricate Gate scans HTML, Markdown, and text files for numeric claims before publication. You supply a manifest containing each allowed value, its claim, and its source URL; unmatched figures fail the command, while malformed manifests fail closed. The tool is read-only, local, dependency-free, and does not access credentials or the network.

Skyll exposes a REST API and MCP server that searches multiple skill sources, ranks matches, and returns complete SKILL.md content with optional references. Agents can choose and load instructions during a task instead of relying on preinstalled skills. A hosted service is available, or teams can self-host the server for more control.

TarkovMonitor is a Windows companion app that watches Escape from Tarkov log files and turns recognized events into audio alerts and visual timers, including match found, raid start, run-through time elapsed, and Scav cooldown. With a Tarkov Tracker token, it can mark newly completed quests; with a Tarkov.dev remote code, it can open the current map and place a screenshot-derived position on it. Basic alerts work on their own, while the broader integrations require the corresponding site token or remote code.