Agent Skills

---

name: website-rebuild

description: 1:1 rebuild of award-winning creative websites (WebGL / scroll-animation / portfolio sites). Evidence-driven pipeline - mirror-first forensics, line-number-traceable reverse engineering of minified bundles, verbatim porting, quantitative verification gates. Use when user asks to "复刻网站", "重建网站", "1:1 rebuild", "clone this site", or provides a URL of a creative/award site to reproduce.

---

name: short-drama

---

name: xhs-visual-director

description: Use this skill when planning, redesigning, or reviewing Xiaohongshu carousel visuals, covers, 3:4 image posts, page-by-page visual direction, image-generation prompts, style selection, layout critique, or XHS captions. It acts as a senior visual director for 小红书图文, not a generic copywriting assistant.

---

name: video-information-reading

description: 从微信聊天记录导出文件中找出视频链接或本地视频附件，整理可获得的逐字稿、内容摘要、关键结论和时间点，并保留来源。当用户要求解读、总结或归档聊天记录里的视频内容时使用。

---

name: wechat-article-extract

description: 从微信聊天记录导出文件中识别公众号文章链接或分享卡片，提取标题、公众号名称、发布时间、正文和图片，并保留原文链接。当用户要求整理、摘录或归档聊天记录里的公众号文章时使用。

---

name: deserialize

description: Insecure-deserialization playbook — fingerprint the language/format (Java serialized, .NET BinaryFormatter, Python pickle, PHP unserialize, Node serialize, YAML/JSON-with-types), then build a working gadget chain with ysoserial / ysoserial.net / phpggc / custom pickle. Use when you see serialized blobs (rO0/AC ED, base64 ViewState, PHP O:) or a parameter/cookie that deserializes user input.

---

name: graphql

description: GraphQL pentest playbook — find the endpoint, dump the schema (introspection or field-suggestion fallback), then test for authorization gaps, query batching, alias overload, depth-based DoS, and SQLi/NoSQLi in resolver arguments. Use when the target exposes a /graphql endpoint, GraphiQL, Apollo, or accepts GraphQL queries.

---

name: jwt

description: JWT attack playbook — algorithm confusion (alg=none, HS/RS confusion), kid path traversal/SQLi, jku/x5u SSRF, weak HS256 cracking, and embedded JWK trickery. Use when the target uses JWTs for auth (header.payload.signature).

---

name: my-skill

description: One line on what this playbook does, then a "Use when ..." clause so the agent knows when to load it (e.g. "Use when the target exposes X / you see Y in requests"). Max 1024 chars. This description is the ONLY thing the model sees until it loads the skill — make the trigger conditions explicit, since there is no separate `triggers` list.

---

name: race

description: Race condition / TOCTOU playbook — limit overrun (one-time codes used twice, gift cards spent twice), single-packet attack (last-byte sync) to force parallel processing, and state-confusion races (file upload + read, order before payment). Use when timing-sensitive logic could be abused — one-time codes, coupons/gift cards, balance or limit checks, double-spend.

---

name: recon

description: External recon playbook for a web target — subdomain enumeration, live-host probing, tech fingerprinting, and a first pass at content discovery. Use when the user gives you a root domain or apex and wants attack surface mapping.

---

name: ssrf

description: Deep-dive SSRF testing — bypass filters, hit cloud metadata, chain to RCE/credential disclosure. Use when a target parameter clearly accepts a URL or hostname.

---

name: ssti

description: Server-Side Template Injection — fingerprint the engine first (Jinja2 / Twig / Velocity / Freemarker / ERB / Smarty / Mako / Handlebars / Pug), then escalate the engine-specific primitive to RCE or sandbox escape. Use when user input is reflected through a template engine (Jinja2/Twig/Velocity/Freemarker/ERB/Smarty/Mako/Handlebars/Pug) or {{7*7}} evaluates to 49.

---

name: takeover

description: Subdomain takeover playbook — sweep subdomains for dangling CNAMEs / NS records pointing at unclaimed third-party resources (GitHub Pages, S3, Heroku, Azure, Netlify, Shopify, ...), confirm with the engine's HTTP fingerprint, then prove impact by claiming the resource in scope. Use when enumerating subdomains for dangling CNAME/NS records pointing at unclaimed third-party services.

---

name: webvuln

description: Web vulnerability hunting playbook. Use after recon, when you have specific hosts/endpoints to test for IDOR/BAC, injection, auth flaws, SSRF, and known CVEs. Emphasizes real PoC + concrete impact.

---

name: oil-ui

description: Design, improve, and review interfaces for websites, apps, dashboards, and components: explore distinct design directions, compare styles side by side, and refine visual hierarchy against real screenshots. Delivers comparison pages, design notes, mockups, or working interfaces. Use for new interface design, style comparison, visual polish, screenshot recreation, or interface review. Not for business logic, APIs, builds, or code cleanup without visible UI changes. If oil-ui-pro is installed, use oil-ui-pro instead. 设计、改进和评审网站、App、后台与组件的界面，完成设计方向探索、多风格同屏比较、视觉层级和基于实际画面的迭代；按请求交付对比页、设计说明、设计稿或可运行界面。当用户需要新界面设计、比较不同设计风格、视觉精修、截图还原或界面评审时使用。只负责设计判断，组件归属、数据流、状态正确性和测试等代码实现质量不在范围内。不用于纯业务逻辑、接口、构建部署、无界面行为变化的代码整理、单独绘制普通插画或操作现有网站。已安装 oil-ui-pro 时交给 oil-ui-pro。

---

name: better-codex

description: Behavioral guardrails for Codex coding work based on common user complaints. Use when Codex is asked to implement, modify, debug, review, test, or operate on a codebase and should avoid unsafe scope expansion, stale edits, fake completion, brittle edits, shallow debugging, superficial patch-on fixes, one-off special-case code, short-term design choices that hurt maintainability, blindly following incorrect user assumptions, poor dependency choices, generic product or UI output, over-mocked tests, noisy approvals, verbose status reports, or leaking internal reasoning into project artifacts or user-facing UI.

---

name: wtf

description: Pre-launch and pre-commit audit for vibe coding projects. Use when asked to check whether a project is ready to ship, deploy, merge, or commit, especially for common AI-built app mistakes: broken project structure, committed secrets or cache files, environment variable hygiene, database migrations, ORM/schema drift, unsafe raw SQL, unused legacy code, dead routes/components, weak auth, missing tests, build failures, and deployment footguns.

---

name: openkb

description: Use when the user asks about content in their OpenKB knowledge base — research topics, concepts compiled from their documents, cross-document synthesis — or mentions `openkb`, an `.openkb/` directory, or a `wiki/` tree generated by openkb. The user may invoke you from any working directory; the active KB resolves via `openkb status`. Do NOT use for arbitrary Markdown directories, Obsidian vaults, or documentation sites not built by openkb.

---

name: openkb-deck-editorial

description: Use when the user asks the openkb chat to make a deck / slide presentation / PPT / slides / 演示稿 / 幻灯片 from their compiled KB content. Generates a polished single-file HTML deck in the Editorial Monocle visual direction (warm cream background, serif type, brick-red accent) — designed to be opened in a browser, full-screened, and shared. Does NOT apply to generating skills (that's `openkb skill new`), long-form research reports, or interactive prototypes.

---

name: openkb-deck-neon

description: Use when the user asks the openkb chat to make a deck / slide presentation / PPT / slides / 演示稿 / 幻灯片 from their compiled KB content AND wants a dark, high-tech, neon / glow / glassmorphism look (赛博 / 科技风 / 暗色 / 霓虹 / 炫酷). Generates a polished single-file HTML deck in the Aurora Glass visual direction (near-black background, teal/sky/magenta/amber neon accents, glassmorphism panels, aurora gradient atmosphere) — opened in a browser, full-screened, shared. For the warm, printed, serif look use openkb-deck-editorial instead. Does NOT apply to generating skills (`openkb skill new`), research reports, or scrolling long-form documents.

---

name: openkb-html-critic

description: Use to review a generated HTML deck or single-page artifact for visual quality and structural correctness. Especially good at catching CSS specificity bugs where slide-modifier classes (.divider, .center, .q, .flow etc.) accidentally override the base .slide{display:none} and cause one slide to stack on top of every other. Also catches missing keyboard navigation, bullet-dump / wall-of-text failure modes, broken self-containment (external link/script/img). Patches the file in place; never changes the original content (slide text, numbers, named entities are the author's work, not yours).

---

name: airtable

description: List Airtable bases, read table records, and add records. Trigger phrases: airtable, my base, add a record, spreadsheet.

---

name: alphavantage

description: Stock quotes and daily price history. Trigger phrases: stock quote, alphavantage, stock price.