Agent skill for behavior-preserving clean-code refactoring of any codebase, based on Fowler's catalog (refactoring.guru): detect smell -> choose technique -> apply in safe small steps.

Router-first engineering skills for AI coding agents: deliberate refactoring, architectural hygiene, ADRs, and bounded multi-language tooling.

backlinks-expert - Claude Code Agent Skill (target LLM: Claude). Part of the mr-bridge.com toolkit.

Human Writer (Hindi (Devanagari)) - AI-text humanizer & 0-100 detector (Claude Code skill). Part of the mr-bridge.com toolkit.

Markdown → one .paged.html print document that edits itself in the browser → PDF via print. Ships agent skills.

Open-source AI agent skills and playbooks for Codex, Claude, Cursor, and LLM workflows

A closed loop of three agent skills for website design: analyze a site, recreate its design system as a static-site template, and audit the build against the original analysis.

the wiki that doesn't trust itself. documents go in, claims get checked against their sources, contradictions get flagged instead of buried.

Two portable agent skills (SKILL.md format) for Hebrew/Arabic RTL work: authoring new RTL content (HTML/PDF/Markdown) and translating existing RTL PDF forms to LTR in place. Usable by any AI agent.

REM Sleep — an AI-agent skill for memory consolidation, learning, and emotional regulation. The inward counterpart to octopus investigation.

Claude Code skill: a phased Production Readiness Review go/no-go gate (reliability, security, operability, cost, observability) with static code probes and a dated PRR record.

Agent skills for LightX2V — teach Cursor and AI agents to generate images & videos (t2v, i2v, t2i, s2v) via the lightx2v CLI and x2v.light-ai.top OpenAPI.

Sunset — retire a dead project with dignity: harvest the lessons, archive with a marked grave (reversible), release the attention-rent. The grief organ alongside octopus + rem-sleep + immune-check.

Agent skills for running security scans — secret scanning (gitleaks), dependency CVE audit (npm/cargo/osv), SAST (semgrep), and a composed pre-release gate. Grounded SKILL.md runbooks for Vanta / Claude Code / any skill-aware agent. MIT.

Skills, hooks, and CLI checks for safer biomedical AI coding agents.

A founder/indie-hacker skills pack for Claude Code, Cursor, Codex and Gemini CLI: market research, competitor analysis, outcome-based pricing, landing-page copy, MVP spec, cold outreach, growth experiments and fundraising narrative.

A Claude Code skill that forces grounded research + a cheapest-falsifying test before Claude commits to any consequential claim. Red-team your own belief.

27 reference-grade Claude Code skills extracting Edward Tufte's visual design principles from all 5 books — data-ink ratio, Lie Factor, chartjunk, small multiples, sparklines, analytical design, and more

A portable discovery methodology for agent skills. Teaches agents how to find skills from catalogs, marketplaces, and GitHub.

Home / DIY omakase planning skill for AI agents — menu, safe fish sourcing, portions, and a time-reversed prep timeline. npx skills add zwbao/home-omakase-skill

Structured agentic development, shipped as Agent Skills - plan, triage, build, and review with autonomous sub-agents, human approval gates, and automatic review on every diff. Runs in Claude Code, Cursor, Codex, Gemini, Mistral, pi, and any Agent Skills harness.

The broadest agent-skills library: 64 behaviorally-tested engineering skills + a 38-role SDLC & marketing org + an orchestrated build loop. Works with Claude Code, Cursor, Gemini, Copilot, Codex.

Erstellt deutschsprachige Mondphasen- und Lunar-Kalender für Monate, Jahre, Journaling, Rituale, Gartenplanung oder redaktionelle Inhalte, inklusive klarer Struktur, Zeitzonenhinweis und vorsichtig formulierter Impulse.

Lightweight intent refinement for AI coding agents. Detects ambiguity before the agent acts — asks at most one question.

Review agent setup files before you run them.

Claude Code skill: audit products for Croatian GDPR compliance (AZOP) and produce interactive HTML audit reports

Multi-engine SEO Website Builder Agent Skill for Google, Bing, Yandex, Pinterest, Shopify, Astro, local business, and AI search.

Standalone WorkBuddy marketplace for VOC-JTBD-NEO-Analysis

apify-scraper-builder - Claude Code Agent Skill (target LLM: Claude). Part of the mr-bridge.com toolkit.

35+ Agent Skills for Claude Code: a full idea-to-ship engineering flow + an original security suite (PII encryption, IDOR/XSS code audits, pentest).

Agentic audio transcription in Go: parallel transcription agents, an LLM judge agent that picks the best, extensible via Agent Skills.

POC Driven Development — build complex features with AI agents by owning one self-contained file per feature and closing gaps until it's ready to implement. An installable Open Agent Skill (/poc).

Beginner-friendly bilingual Codex skill for finding verified photography contests through guided choices

Agent handoff skill for saving task state and continuation plans

干活 AI IP 质量筛选 Skill：批量检测代理出口、ASN/ISP/地区一致性、同段集中和网络环境信号，输出可复核排序表

Token-efficient shared project memory skills for developers and AI coding agents.

Test bench for custom developer-agent skills: SQL schema validation, Conventional Commits formatting, JSON-to-Pydantic conversion, and license-header injection.

Agent skill for migrating Next.js 13–16 (Pages or App Router) to Astro 7. Plan scrutiny, route mapping, Astro-first components, minimal React islands. Install with npx skills add.

Portable file-based knowledge format with validation, indexing, auditing, and migration tooling.

A verification discipline for AI assistants: ground every load-bearing claim in a real source consulted now, or abstain. Community-first, never training memory. Origin: the Vietnamese discipline xac tin.

Agent skill to integrate Graphify knowledge graph into any project

Codex skill to clean lecture transcripts and generate courseware-aligned study notes with Mermaid mindmaps and optional Feishu/Lark writeback.

Immune Check — a pre-flight reflex that scans any outbound artifact for secrets/PII before it leaves the body. Stack-agnostic. The defense organ alongside octopus + rem-sleep.

Agent skill for implementing WebMCP on websites — expose app features to browser agents via document.modelContext.registerTool(). Install: npx skills add Blackie360/webmcp-skill -g

Maintains ONE canonical asset as the source of truth and derives every channel atom — blog, newsletter, LinkedIn, X thread, short-video script, email — from it, with traceability and brand-voice gates so nothing drifts.

Runs a documentation site like a product: audience & IA → Diataxis-structured docs-as-code repo → auto-generated API reference → CI doc-drift gate → publish, with verification at every step.

Skills anti-marketing / décroissance pour Claude

An open Agent Skills + stdlib-Python toolkit that turns messy bank/app financial exports into one clean, de-duplicated, categorized dataset and a private local dashboard. Works with any AI coding agent — or standalone. Runs entirely locally.

Ditsmod Agent Skills

Your AI agent is a yes-man. This fixes it. No empty praise, no hype, no 'You're absolutely right!'. Works in Claude Code, Cursor, Codex, Copilot, Windsurf, Cline, Kiro, and more.

⚛️ 18 production-grade AI agent skills for the React ecosystem — install into Claude Code, Codex, Cursor, OpenCode, Pi, Kiro & more. By Pythoughts.

Agent Skill for careful talking-head video cleanup, HyperFrames motion planning, preview confirmation, and final MP4 export.

Agent Kaizen skill for Claude Code & Codex: faithful, identifier-preserving reference for DaVinci Resolve (Blackmagic Design), Media, Cut, Edit, Fusion, Color, Fairlight & Deliver pages; timelines, trimming, retiming, color grading (RCM/ACES/LUTs), Fusion compositing & tracking, Fairlight audio, Resolve FX, and rendering.

Anti-slop Claude Code skills: ground Claude in measured reality — live GitHub SOTA ranking + real design-token extraction — instead of training-data priors.

Provide guidelines on how to develop Ansible playbooks that are compliant with project standards. Should be used when generating, modifying, or reviewing Ansible playbooks, roles, or collections.

A monorepo of agent-agnostic AI skills for Claude Code, Codex, Gemini CLI, Devin, and more

Claude Agent skill: find a field's real core people by mapping its endorsement graph — following-BFS + cross-source ranking on GitHub, OpenAlex, or any platform. Public data only.

Provide guidelines on how to determine appropriate git commit messages that are compliant with project standards. It should be used when generating, modifying, or reviewing git commit messages.