Replace a brand logo printed on an existing product photo and finish the result as clean studio product photography. Runs fully locally — no API keys, no paid services.

Ranks unresolved questions by expected loss so the highest-risk unknowns are addressed first.

Quality-first, fully local PDF → Markdown OCR for large batches of poorly scanned documents. Any Tesseract language, especially strong on German. Ships as a CLI tool and as an agent skill for Claude & ChatGPT/Codex.

开源中文去 AI 腔与个人风格润色 Skill：锁定事实、立场和作者声音，减少模板化表达。

AI design-critique skill that teaches measurable craft and evaluates taste in context

A Claude skill that turns a production AI agent incident into a golden eval case. The loop that feeds the two gates.

A Claude Code skill to audit and sync project context (CLAUDE.md, skills, project memory, and settings). Keeps your AI assistant fully aligned with the codebase at the start of new sessions or after major changes.

Compile fuzzy findings into evidence-backed, proof-gated issues that parallel coding agents can ship safely. Agent Skills format for Claude Code, Codex, Cursor, and Markdown-native agents.

Governed observability + governance for on-endpoint local LLMs (Ollama): model policy, deterministic prompt scanner, route-through guarding — the complement to IGEL AI Armor. 18 MCP tools (preview)

A portable Agent Skill for evidence-first project defense review, follow-up questioning, answer scoring, and material repair.

Uses transaction reconciliation to verify payment claims across the processor, ledger, idempotency records, and settlement data.

Quizzes developers on recent code changes and grades their answers, using retrieval practice to surface comprehension debt.

The markdown doctor for AI-written code. Diagnose, treat, and prevent AI slop with any coding agent.

바이브 코딩에 규율을 입히는 에이전트 스킬 모음 · Agent skills that bring discipline to vibe coding. Install: npx skills add sjsylee/skills-hub

Agent Skills for safe local Numera workbook automation.

Agent Skill for AI coding agents: integrate Countly analytics with a proper data collection model (Claude Code, Cursor, Codex, Copilot, any LLM)

Master accordion. N parallel sessions → compress each, diff, wake one, seal the rest. Cogpros primitive 29.

Command-line tool for AI agents (Claude Code, OpenClaw, Cursor) to draft, schedule, and publish social posts through posthell. Agents draft by default; publishing to non-X networks is opt-in per key.

开源 AI 岗位匹配与简历优化 Skill：分析 JD、建立证据矩阵、判断是否值得申请。

Agent Skill that turns terse intent into structured, verifiable prompts

Recover Codex side chats, export them to Markdown, and continue them as durable tasks.

可能AI最大的价值就是给你泼冷水，让 AI 停止默认迎合，执行证据化反方审查：泼冷水、红队、证据审计、premortem 与决策复盘。

Official MailKite skills for coding agents — npx skills add mailkite/agent-skills. Email for every product you ship: receive as a webhook, send over a verified domain, give an agent its own inbox.

The method behind my Candidate Go-to-Market pipeline — written judgment, failure log, and evidence ledger of a job search run as a live GTM motion.

HA2HA protocol and the hosted MDSync human-agent collaboration workspace

Barbara Minto's complete Pyramid Principle as an executable, corpus-verified skill — ten hard-stop working sessions, dogfooded README

Agent skill: RFQ and supplier quote comparison for procurement — structure RFQs, collect quotes, L1/L2/L3 ranking, place purchase orders via the AuraVMS API. Works with Claude Code, OpenClaw, Codex.

Lightweight skill for preserving and coordinating project state.

Agent skill for bounded multi-agent strategy selection, dispatch, durable state, review gates, and recovery.

Agent skill: safe re-initialization of project agent knowledge — salvage-verify old docs, backup-gated deletion, fresh codebase read, direction interview, rebuild cited AGENTS.md/ARCHITECTURE.md/ROADMAP.md

📖 使用 Agent Skill 生成忠于原著的故事梗概，并通过书脉网站提供三种阅读深度的连续阅读体验。

See what your AI agent actually knows — an interactive map of your plugins, skills, hooks, and MCP servers, what references what, and what fires on a prompt. Zero dependencies.

Agent Skills that teach Claude to build Rails 8.1 apps — vanilla-first stack, pure RSpec, Hotwire (Turbo/Stimulus/Native), OpenAPI via rswag, ruby_llm AI, Kamal 2 deploys

Public agent skills for Jira-backed development workflows

Battle-tested skills for Claude Code and other coding agents — distilled from a real production automation setup

Open collection of practical skills for AI agents / AI Agent Skills 开源集合

HeyEddi Cursor Agent Skills hub — 20 skills, agent eval suite, git subtree distribution

Bilingual Codex skill for AI product idea brainstorming, adversarial pressure testing, validation, and concept briefs.

讓 AI 陪你練判斷的繁中 skill：校準式五維判斷框架，先檢驗後表態、不附和也不硬反對，裁決權交還給你。適用 Claude Code / Codex / 任何 AI 工具。

Agent skills for deliberate, efficient AI-assisted work

The working discipline of Claude Fable 5 as installable skills — method, taste, and judgment for any Claude model, with a built-in eval suite that measures the delta.

Agent Skill for page-cited research talk packs from academic PDFs.

Turn authorized creator profiles into source-linked content libraries with coverage evidence, recurring patterns, topic summaries, and breakout analysis.

Agent skills for Sonar — App Store Optimization for AI agents (keyword research, ASO audits, revenue estimates)

Free AI quote-readiness checklist for custom cabinetry and built-in woodwork

Open-source factory for generating, validating, and publishing reusable Agent Skills

One Linux Agent Skill for package errors, CVE/CNNVD, distro migration, ELF/ABI, command portability, and containers, powered by PkgSeek.

Agent Skill for authorized passive security scans of AI-built and vibe-coded websites

A judgment layer that helps AI agents build the right thing, ask fewer better questions, and preserve purpose through consequential work.

Codex skill for a fresh critical second-opinion review, fix, and re-review loop.

PASS/STOP pre-action receipts for Codex and AI coding agents before edits, handoffs, commits, or publish steps.

Cross-platform agent-to-agent comms: an append-only in-repo mailbox where AI coding sessions coordinate — and grill each other's handoffs for evidence.

Agent Skills

Skills that design the work: a growing toolkit of disciplined, harness-agnostic processes for coding agents.

An operations-analysis method for AI agents: skills, role agents, data-hygiene rules and artifact templates - OTIF audits, forecast reviews, ABC-XYZ, safety stock, Pareto discipline, reporting contracts. Pitfalls included.

Portable agent skills featuring Agent Workflow 1.0: native parallel teams, evidence-based review, and fresh-context verification.

netriage — 先分诊、后开方：给 AI agent（Claude Code / Codex）用的 Linux VPS TCP/网络循证调优 Skill。Evidence-based VPS TCP tuning: question, inspect, test, recommend, apply only after approval.

🤖 X(Twitter) 热点自动转微信公众号爆款文章 — AI 生成去AI味长文 + browser-use 自动化发布到草稿箱

Skills for Claude Code: Serena MCP onboarding and memory refresh, plus terminal capture with asciinema

Agent skill: Indian compliance deadlines — ROC, GST, TDS, PF/ESI calendars, penalties, and government scheme matching. Works with Claude Code, OpenClaw, Codex.