Agent skills (open standard) for the bookends of the agent loop: context assembly, decision review, and work-shaping outputs. Works with Claude Code and any Agent Skills harness. Ships its own adoption and value measurement.

A compact AI coding-agent skill for designing, implementing, and reviewing boringly reliable backend APIs.

Portable agent skill for designing roguelike and roguelite games, systems, levels, balance, progression, and procedural generation.

ROS2 Jazzy + Gazebo Harmonic simulation expert skill for Claude Code — build, run, verify, and analyze robot simulations end to end

A strict 5+1+2 agent workflow that turns ambiguous project ideas into verified delivery.

A vendor-neutral preflight protocol for AI agents to frame goals, risks, boundaries, and success before acting.

Production-oriented frontend design Agent Skill with zh-Hant, true mobile UX, evidence-bounded model routing, and Playwright evals.

Slack-ready end-of-day reports from git commits for AI agents

Claude Code skill: follow YouTube channels, fetch transcripts (free captions with Whisper fallback), and write daily digest briefs. Zero API keys required.

This Repository showcases a SKILL for agentic models like Claude Code, Antigravity, Cursor, Kiro, and others and is capable of turning a simple react application into a native windows application with its own custom-made, modern installer.

Claude/Agent Skill that audits any URL for AI-search readiness. Checks llms.txt, robots.txt AI-crawler rules, JSON-LD schema, and citability patterns.,

Give this repo to Claude and it becomes your training coach: real periodized plans, Strava + Apple Health data, progress tracking and grading — files are the database, skills are the features, the agent is the app.

Discovery-only Agent Skill and CLI for finding, filtering, and ranking bug bounty programs with safe scope-aware audit handoff.

Nobody asked for this. Nobody fully understands it. Six Claude skills for web design registers, plus a router. Named at 3am, no regrets.

Video-game-like skill trees for tracking real growth - framework-agnostic engine, self-hostable app, and agent playbooks

A team of domain experts as installable agent skills + subagents for Claude Code, Codex, and Pi. Each expert answers from a curated, cited knowledge folder - no RAG, no vector DB, no embeddings.

Second-opinion code review: a different model reviews each diff, and disagreements go to a human. Multi-model deliberation.

A single-file, zero-dependency CLI that gives coding agents durable memory — reads/writes a .brain/ harness (features, checkpoints, rules, recipes, plan reviews) as token-efficient TOON.

An operating system for engineering: durable, opinionated playbooks (Agent Skills) for software architecture, AI/agentic systems, product engineering, and technical leadership. Read by humans, loaded by any LLM/agent.

GitHub-distributed Codex plugin with practical, evidence-driven Agent Skills for software engineering and legacy project onboarding.

Governed GPU inference ops (vLLM + Ray Serve): latency RCA, scaling, drain, 30 MCP tools (preview)

Symptom-indexed Codex CLI/App troubleshooting skill knowledge base

Hire an agent, not a prompt. Give it a business outcome; it diagnoses your system from real data, finds the constraint, and builds self-improving agents with evals and graduated autonomy. Runs in Claude Code and Codex.

A personal learning OS for Claude Code + Obsidian: domain-agnostic course/paper ingest framework with verifiable feedback loops. 通用学习操作系统框架

Claude Code plugin: linked project notes for Obsidian with auto vault detection

Agent Skill: teach AI coding agents (Claude Code, Codex, Gemini CLI...) to edit video in DaVinci Resolve via its native Python scripting API

Steal the mood, not the pixels — a Claude Code skill that reads the design DNA of sites and repos you admire, then writes a version that fits your project.

Schedule background alarms using natural language that bypass the 15-minute native agent cap.

Interviews you before you build. Structured, one decision at a time - from problem to architecture - before your coding agent's plan mode takes over.

Give Claude and other AI agents eyes: watch and transcribe videos from YouTube, Instagram, TikTok, and more, frame by frame.

A skill for creating technical documentation and guides with Astro and distill.pub-inspired styles.

Uses cost-aware LLM routing based on total cost, including corrections and review.

A practical framework and AI skill for critiquing agentic experiences, oversight, trust, and consent.

Evidence-first safety, drift detection, and verification guardrail for AI agents

A lightweight Agent Skill for initializing and maintaining AGENTS.md and repository docs.

用金庸式命名方法，为真实生活生成常用、顺口、有文化根基的中文姓名 Agent Skill

8 AI学习诊断工具 — 卡在哪用哪个，不卡不用。Hermes/Claude Code/Codex/Cursor/Trae 五端兼容。

Agent Skill to extract research-paper figures, captions, pages, and provenance manifests.

Agent skills for Claude Code: research that starts before the sessions, and catching visual drift in a rebuild.

Agent skill: evidence-based feasibility + migration strategy for architecture transformations (language ports, standalone→SaaS, vanilla→framework) — dependency census, difficulty heatmap, GO/PARTIAL/NO-GO verdict, parity-gated roadmap

Open-source agent skills & tools from Growth Garage. First up: kb — a knowledge base for AI agents that maintains itself.

Agent skills for sourcing startup deals with Frontrun — pre-round signal from 1,000+ investor follow graphs on X. For Claude Code, Cursor, and any MCP-capable agent.

Run nanoodle workflows from Python — zero-dependency executor for saved noodle-graph.json files

Local-first health and fitness Agent Skill for Hermes, WeChat, Google Health, and Excel

Autonomous issue-to-clean-PR skill for Claude Code and Codex: plan, validate, implement, self-review, ship, and monitor CI/review feedback until clean

Convert research paper PDFs into structured Obsidian notes with automated translations, figure analysis, and concept explanations for deep study.

CRISP-DM as 7 sequential agent skills for OpenAI Codex CLI — data-pipeline projects, no frontend. Codex port of crisp-for-data + earn-the-data.

Multi-agent skill development pipeline for Claude Code: plan → develop → 9-layer test → fix, with a context-blind orchestrator

Thinking lenses for Claude Code — a thought-out answer instead of a quick, generic one.

Public cross-agent skill for finding a commercially useful book topic, reader, positioning angle, and client-attraction path.

Experiential learning pipeline: read a programming book, implement, verify by test (mutation-checked), and accumulate verified skills.

Evidence-backed Seedance 2.0 prompt engineering skill for Codex: Dreamina, BytePlus, Volcengine, Fal, multimodal references, editing, and prompt linting.

A Claude Agent Skill that sharpens rough prompts using Anthropic's own prompt-engineering guidance.

Lower LLM costs for coding agents: delegate implementation to cheaper models and reserve frontier-model tokens for evidence-backed final review.

Verified CONTINUE.md checkpoints for Codex and Claude Code sessions

Seven read-only audit skills for AI coding agents: codeauditor, secauditor, dbauditor, llmauditor, seoauditor, uiauditor, uxauditor. Each scores one dimension of a codebase end to end and writes a prioritized report. One repo, one install. Claude Code, Codex, Cursor, and any Agent Skills harness.

Portable workflows, templates, and Agent Skills for tuiweave applications

[ENG]Outputs a structured YAML format to explain the situation to other AIs when stuck on an implementation issue. / [JPN]実装に行き詰まった際、他のAIに状況を伝えるための構造化されたYAMLフォーマットを出力します。

Layered diagnostic-and-repair skill for Chinese / non-ASCII encoding problems on Windows — for AI coding agents (Claude Code, Codex, any Agent Skills runtime).