Appllama MCP for mobile design research: search top-grossing mobile apps, walk their screens, study onboarding and paywalls, then build from what wins.
AI-powered blog skill suite: 32 skill directories (1 orchestrator + 31 sub-skills) and 5 agents to write, rewrite, audit, and optimize content for Google rankings and AI citations (GEO/AEO). Includes the FLOW framework, topic clusters, multilingual publishing, Google API integration, personas, fact-checking…
Brief-led image generation, editing, visual review, and bounded multi-model orchestration with current Google Gemini image models. Requires Claude Code 2.1.199+ and Python 3.11+ with python3 on PATH.
★not rated 1.1k▲
+15 7d agoA
tokens not measured
originalMIT
Cyber threat intelligence and OSINT toolkit — structured investigations with sourced, trust-scored findings. Bundles the skill, the /cti command set, the typed intel MCP surface, and the RULE 1 / outbound-action safety hooks as one installable unit.
★not rated 611▲
+3
changed 1mo agoA
tokens not measured
AI-orchestrated visual production for Obsidian Canvas. Create presentations, flowcharts, mood boards, knowledge graphs, and galleries with intelligent layout and AI-generated content.
★not rated 301▲
+2 6mo agoA
tokens not measured
originalMIT
Open-source cybersecurity analysis agent. Scans any local project for vulnerabilities: code security (SAST), dependency CVEs (SCA), secret leaks, authentication/authorization flaws, cryptographic weaknesses, misconfigurations, supply chain risks, and CI/CD security. Covers all OWASP 2025 Top 10 and CWE Top 25…
★not rated 285▲
+2 3mo agoA
tokens not measured
originalMIT
Detects 55 AI writing patterns and rewrites text in five voice profiles so it reads like a specific human wrote it, with an optional 0-100 AI-tell score. Use when text sounds AI-generated or like a chatbot, when preparing a blog post, README, or LinkedIn post for publication, when auditing prose for AI tells, or when…
This skill should be used when the user asks to "upload images to Cloudflare", "implement direct creator upload", "configure image transformations", "optimize WebP/AVIF", "create image variants", "generate signed URLs", "add image watermarks", "integrate with Next.js/Remix", "configure webhooks", "debug CORS errors"…
Interactive longform-to-shortform video creator. Extracts viral-ready short clips from long videos with Remotion-rendered animated captions, AI segment scoring, and platform-optimized exports for YouTube Shorts, TikTok, and Instagram Reels.
★not rated 226▲
+8 3mo agoA
tokens not measured
originalMIT
Grand Slam Offer framework. Skills and agents for market research, offer building, value optimization, pricing, positioning, hooks, and landing page copy — based on Alex Hormozi's methodology.
★not rated 216▲
+15 10d agoA
tokens not measured
originalMIT
Local-only skill quality and security evaluator for Claude Code/OpenClaw. Scores six dimensions, runs local validators, and stores reports and snapshots under .skill-compass/.
★not rated 215▲
+1 5mo agoA
tokens not measured
originalMIT
Bypass bot detection using rebrowser-playwright (Node.js) or undetected-chromedriver (Python). Passes bot.sannysoft.com fingerprint checks and automates Google without triggering CAPTCHA. Authorized-use only — respect each site's ToS.
★not rated 203▲
+2 1mo agoA
tokens not measured
originalMIT
Ultimate Claude Code skill creator. Design, scaffold, build, review, evolve, and publish production-grade AI agent skills following the Agent Skills open standard and 3-layer architecture.
★not rated 182▲
+3 6mo agoA
tokens not measured
originalMIT
Content Repurposing Engine for Claude Code. Transforms any content (YouTube videos, blog posts, podcasts, local files) into 17+ platform-optimized outputs for Twitter/X, Threads, LinkedIn, Instagram, TikTok, Pinterest, Snapchat, Facebook, YouTube Community, Skool, Discord, Reddit, Quora, Medium, WhatsApp Channels…
★not rated 160▲
+3 6mo agoA
tokens not measured
originalMIT
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: