Recently updated open-source projects

local-flywheel-wp-testing uses Local by Flywheel’s bundled mysql client and per-site Unix socket when an external wp-cli cannot reach the database. It seeds data through SQL, verifies the resulting pages over HTTP, and accounts for serialized metadata, page_id redirects, changing run identifiers, and cleanup. The skill is aimed at testing plugins or themes against a Local-managed site, so it requires that local environment and should not be treated as a general WordPress database tool.

claude-skill-univer provides a dependency-free Python command-line tool for Obsidian Univer .univer.md spreadsheets. It reads and writes cells or ranges, clears values, lists sheets, adjusts column widths, freezes rows, and applies colors or borders using A1-style references. Numeric values and formulas are detected automatically, and a sheet can be selected explicitly. The current scope does not include several text styles such as bold, italic, underline, or font family changes.

The skill maps a repository’s technologies and attack surfaces, runs a bundled set of SAST, dependency, secret, container, and infrastructure scanners, then performs parallel deep dives into areas such as authorization, injection, supply chain, and agentic risks. It emits Markdown, SARIF, CycloneDX, and a reusable baseline, supports Claude Code only, and is best run inside an isolated container.

video-editor combines separate screen or camera takes into one editable Screen Studio project, then makes editorial cuts that remove dead air, filler, and redundancy while preserving useful pauses. It can clean and level the voice track, add a coded branded intro, and export either an editable project or a finished MP4. The skill targets macOS, uses ffmpeg, and relies on Screen Studio’s open project format rather than performing only mechanical silence trimming.

AgentSilex is a roughly 300-line Python framework built around readable Agent, Runner, Session, and tool abstractions. LiteLLM lets it switch among more than 100 model providers, while the framework adds tool calls, session history, multi-agent handoffs, MCP clients, streaming, and structured output. OpenTelemetry tracing is included for inspecting execution, and the package requires Python 3.12 or newer.

claude-llm-wiki turns articles, PDFs, notes, and transcripts into a linked Markdown knowledge base organized by the domains you choose. It creates bridge articles between domains, answers questions with citations, checks broken or asymmetric links and contradictions, and exports a compact context layer for another assistant. Raw sources remain untouched, while the wiki, outputs, and session records stay visible in a Git repository that can also be opened as an Obsidian vault.

Evidentia extracts DOI, PMID, arXiv, and NCT identifiers from medical writing and checks them against services including CrossRef, PubMed, OpenAlex, arXiv, and ClinicalTrials.gov. The CLI and MCP server classify citations as verified, bibliographically mismatched, hallucinated, or needing content review; the Claude Code skill adds a 15-criterion appraisal of how evidence is used. It is for writers, editors, and researchers, not clinical decision support.

github-growth-kit reads a repository’s project materials and open issues before creating seven specific community prompts, including a feature demo, Show and Tell thread, milestone tracker, article idea, and contribution path. It avoids duplicate topics and offers a separate command to draft and post a welcoming reply when someone volunteers to help. The skill targets Claude Code, Cowork, and Claude.ai, and is intended for repositories that need an initial community conversation.

shap-figures provides a reusable Python library for publication-oriented SHAP, GAM, and GIS visualizations. It includes coolwarm beeswarms, feature-importance charts, GAM dependence plots with confidence intervals, predicted-versus-actual views, spatial SHAP maps, trends, ablation grids, and labelled multi-panel composites. The package uses a consistent visual style and can be called directly from Python. NumPy, pandas, Matplotlib, and SciPy are required; mapping and image-stacking features add optional dependencies.

github-launch-agent divides a GitHub launch among parallel agents that inspect the project, research comparable repositories, score naming and search choices, generate a social preview, and prepare README and release files. It can also set up GitHub Pages and produce a seven-day distribution calendar with ready-to-use copy. The skill is designed for Claude Code, Cowork, and Claude.ai, and trades a long sequential checklist for coordinated agents and a shorter launch session.

seed-discussions inspects the project’s documentation and package files, identifies its type, and creates five tailored GitHub Discussion posts covering a roadmap, a vote, Show and Tell, usage tips, and troubleshooting. It can enable Discussions through the GitHub API and publish the posts with GraphQL, while leaving the final pinning of the roadmap thread to a person. An authenticated gh CLI and permission to modify the repository are required.

elevate-ui applies a repeatable visual pass to flat, generic web pages so they feel designed rather than AI-generated. It works across light, depth, and responsive life, replacing emoji with bespoke SVG icons and adding restrained texture, glass-like layering, cursor response, and one purposeful animation. The recipes are self-contained inline SVG, CSS, and JavaScript, need no CDN, and honor prefers-reduced-motion.

Enter a date to collect public ETF fund-flow data and receive a structured view of broad-market direction, sector rotation, cross-border sentiment, and an overall buy, sell, or hold signal. It is designed for investors and quantitative researchers using flows to confirm trends rather than predict them, and its output is research material rather than investment advice.

solana-token-safety-skill gives Solana agents a market-integrity second opinion before they trade or launch a token. It examines liquidity, holders, quotes, volume, and flow signals for fake market caps, wash activity, price bait, or a concentrated single-wallet float, then returns a block, caution, ok, or unknown verdict with human-readable reasons.

UGC Video Skill writes prompts for short videos that feel filmed by a real person on a phone, with handheld movement, natural light, believable skin, and casual dialogue rather than polished advertising. Its renderer targets official APIs for Seedance, Kling, Veo, Sora, and Hailuo, adapting reference images and native audio where supported. Provider credentials come from environment variables, and each model has different parameters, input rules, and dependencies.

ui-ux-doctor scans React source and FastAPI code with only Python’s standard library, flagging dead controls, rendering problems, accessibility gaps, and integration mistakes such as unsafe CORS. A static mode reads source files, while a live mode uses an already installed browser to inspect the rendered DOM, contrast, labels, dialogs, console warnings, and failed requests. It needs no pip, npm, or network access, making it useful on locked-down machines.

spark-forge packages local ComfyUI generation on NVIDIA DGX Spark as Claude Code plugins for images, 3D models, music, video, and gated publishing. FLUX.2, Hunyuan3D, ACE-Step, WAN, and HunyuanVideo are selected through model profiles and quantization options, while setup skills provision the stack and private S3 or CloudFront delivery. It requires a ComfyUI container, Docker, AWS CLI for publishing, and downloaded model weights. Model licenses vary and some are non-commercial, so the repository’s terms must be checked before use.

GCC Market Intelligence helps B2B and B2G founders prepare for entry into Saudi Arabia, the UAE, Qatar, Bahrain, Kuwait, or Oman. After narrowing the country, buyer type, and company stage, it surfaces relevant procurement mechanics, localization rules, sovereign funds, soft-landing programs, events, and customer-development frameworks. It is not a live database or legal and tax advice, so changing regulations, quotas, and program terms still need current verification.