Built around the parts of API integration everyone skips until 3am: token refresh, pagination to the last page, 429 rate-limit handling, retries with backoff + jitter, idempotency keys so a retry doesn't double-charge, timeouts, and typed error handling. It reads the API's own docs, matches your stack, keeps keys in env, and produces a client that survives the network blinking — not a naive fetch that falls over the first time the API throttles you.
Turn a rough skill idea into a structured blueprint: named, prompted, differentiated, and listing-ready.
Most recipes are calibrated to someone else's kettle. This Agent builds yours: grain bill with percentages, hop schedule with computed Tinseth IBU, yeast and pitch-rate math, water treatment, strike temperature... scaled to your batch size, efficiency and equipment, with every formula shown so you can adapt when the shop is out of Maris Otter. It justifies each decision, checks the design against BJCP ranges, and flags where the beer is most likely to go wrong. Also diagnoses beers that did.
Audit and manage local AI Skills across Codex, Claude Code, and project workspaces. See what is actually active, detect duplicates and variants, compare changes over time, and generate an offline interactive management map. Read-only by default; every real change requires your approval.
用简单的话说出你希望AI编码智能体(Claude Code、Codex、OpenClaw)稳定完成的任务，即可获得一套完整、可直接发布的Skill包：结构规范、带激活优化描述的SKILL.md，可选的配套子智能体定义与hooks片段，一份验证清单，以及大多数作者会跳过的部分——发布前的安全自查，用来揪出密钥泄露、权限过宽和提示词注入风险。这是一套可复用的创作流程，不是固定模板。无需API密钥，不涉及个人数据。
Expert in end-to-end AI-assisted software delivery, from design engineering and design systems to shipping: design engineering, design system automation, technical documentation, spec-driven planning, polyglot code quality, systematic QA, release auditing, LLM context engineering, and Astro frontend optimization. Built for agent-agnostic workflows across the full 9-skill delivery lifecycle.
I specialize in safe Codex skill onboarding. This Agent inspects unclear websites, repositories, archives, file trees, snippets, and marketplace listings before installation. It explains candidate skills, identifies when Codex would use them, flags obvious installation risks, and requires explicit user confirmation before anything is downloaded or added to the skill library.
AI Integration — Plans and builds LLM API connections (OpenAI, Anthropic, and more) into real apps. Covers model selection, cost estimation, key safety, error handling, and working code. Adapts to non-engineers and developers alike. Stops the #1 mistake: shipping your API key to the browser.
Turn prompts or workflows into testable, launch-ready AI Skill products.