A curated collection of reusable agent skills for the Agent Skills ecosystem.
Agent Skills defines an open format for packaging and sharing agent capabilities as portable skills.
Each skill is usually organized as a folder with a SKILL.md file, and can include:
- clear usage instructions
- references and templates
- helper scripts
This makes skills easy to discover, reuse, and version across different agent runtimes.
This repository currently includes the following skills (30):
agent-first-product-strategy: Reframe AI product strategy from human-first assumptions to agent-first positioning, metrics, and execution.agent-teams-guide: Guide for orchestrating and controlling Claude Code agent teams.analogy-commentary: Turn source content into a familiar analogy framework, then produce concise commentary angles and a short take.china-model-selection-guide: Compare China LLMs with practical, scenario-based model selection guidance.claim-ledger: Turn research claims into a verifiable Claim-Evidence-Boundary ledger.codex-deep-research: Run web-first deep research for comparisons, evaluations, market scans, and citation-heavy recommendations.cross-domain-thinking-toolbox: 25 professional mental models for complex problem solving.debog-yourself: Diagnose and escape psychological traps that block progress.deep-productivity: Deep-work productivity framework (Building, Maintenance, Recovery).defrag: Analyze code for refactoring opportunities and suggest the top 10 highest-value improvements.doko-page-reader: Read and extract content from web pages via a locally connected Chrome browser and Dokobot bridge.ears-requirements: Rewrite and validate requirements using EARS syntax.fastnote-cli-operator: Operate the FastNote local notebook through natural language by mapping requests to note CRUD, search, pin, and tag commands.fix-orbstack-docker-pull: Diagnose and fix Docker image pull failures on macOS with OrbStack (proxy, TUN, fake-ip, registry issues).follow-builders-lite: Lightweight AI builders digest from the upstream zarazhangrui/follow-builders GitHub Actions feeds.hn-to-x-poster: Review the current top Hacker News stories, select software or AI items, and publish a Chinese thread of at least 3 posts to X from an already logged-in browser session.hn-top10: Fetch the current top Hacker News stories as structured agent-friendly output.instreet-operator: Operate InStreet forum and Playground workflows through a bundled Python client with restored local account state.learning-first-principles: Diagnose and optimize learning methods with first-principles thinking.longitudinal-cross-sectional-research: Write deep research reports that combine longitudinal history with cross-sectional comparison.multi-agent-systems: Design orchestrator-subagent LLM architectures.mungers-lattice: Multidisciplinary decision analysis with Munger-style mental models.optimize-gpt-5-6-prompts: Review, simplify, migrate, and create prompts for GPT-5.6 and the GPT-5.6 family.research-content-router: Route research-to-content workflows from research through synthesis to publishable output.resume-builder: Generate and revise job resumes from raw notes, existing resumes, career histories, or profile snippets.retrospective-master: Structured retrospective coaching using the GRAI method.saas-agent-toolkit: Turn SaaS capabilities into agent-usable tool contracts with guardrails, approvals, and auditability.url-to-markdown: Convert a public webpage URL into a reusable Markdown file with a bundled script and agent-friendly execution flow.vibe-design: Design, specify, implement, or review AI-native and Agent product interfaces with product intent, domain semantics, GenUI/A2UI controls, and evidence-based evaluation.x-thread-poster: Post multi-part X/Twitter threads through Chrome DevTools MCP using the user's logged-in browser session.
Install this skill pack:
npx skills add https://github.com/hexbee/hello-skills