146 boosters for "claude-skill" — open source, verified from GitHub, ready to install
一个好的行业 master skill 不是行业百科，是可运行的行业认知操作系统： prompts 里两个不同的目录占位符: 工具脚本 (sourceverifier.py / colddetector.py / quality_check.py / collectors / ingest / transcribe) 不复制到生成的行业 skill 目录, agent 直接调 . 行业 skill 只装载 prompts + research notes + SKILL.md + sub-skills + cli/.
Format markdown articles with bm.md styling and publish directly to WeChat Official Account drafts via the official API. 1. Read markdown file (typically from article directory) 2. Upload local images to WeChat CDN via API
Use AskUserQuestion to ask the buyer: Tell the user the version was updated, then re-read the EVALUATION.md file from the updated directory and proceed with the skill. After the preamble, read the full evaluation methodology:
The definitive SEO and Generative Engine Optimization agent. LLM-agnostic — works on any platform that reads . Merges Google's official SEO guidance, 2026 GEO research, and practitioner best practices into one universal framework. Every finding comes with a
你精通 LaTeX/TikZ 绘图和 draw.io XML 生成，擅长将论文中的系统架构、协议流程、技术方案、技术路线图转化为高质量配图。 执行任务时不要陷入惯性——不是所有图都该用 TikZ，不是所有架构都该自下而上，不是遇到编译错误就反复微调同一行代码。带着目标进入，边画边判断，遇到问题就诊断根因，发现方向错了就换方案——全程围绕「这张图要传达什么信息」做决策。 ① 定义成功标准：这张图要传达什么信息？读者看到后应该理解什么？几个模块、几层关系、什么样的视觉层次？这是后续所有判断的锚点。
引导用户完成一次结构化需求访谈，收集配套素材，最终打包生成一份可直接使用的完整Skill文件包。 1. 用户不需要懂Skill格式——他们只需要回答问题，你来负责转化 2. 每轮问完先小结确认——不要一口气把12个问题全抛出来
The captain's log for your codebase. Every decision, discovery, and change logged as you ship code. Use GitHub as a complete knowledge graph where every brainstorm, commit, review, and decision is traceable. This skill orchestrates existing skills and references; it defines when and how to invoke th
Videos with 80%+ completion get 2-3x algorithmic reach. Your hook determines everything. Layer these elements for maximum impact: 1. Base Layer — Ambient bed (room tone, wind, hum) or silence
Seedance 2.0 accepts multi-modal inputs and produces short-form cinematic video output. Before writing any prompt, confirm the assets available: Before generating a prompt, answer these questions: Every course promo follows one arc:
The first two seconds determine whether a viewer stops scrolling. These patterns open with immediate visual authority. Luxury camera work operates within a narrow, controlled vocabulary. No exceptions. The most used luxury camera move. Camera travels toward subject at 0.1x–0.3x of normal speed. Tota
Transformation content is the highest-engagement format on social media. This skill generates precise Seedance 2.0 prompts that capture the full emotional arc of change: tension, reveal, payoff. Before generating prompts, establish: Show both states in rapid alternation — before/after/before — withi
Transform podcast audio into cinematic visual content using Seedance 2.0 on Higgsfield. This skill produces video prompts that replace static audiograms with storytelling-driven visual experiences built entirely from constructed imagery. The hook is the opening frame that stops the scroll. It must c
The most powerful personal brand camera move. Communicates: "this person is worth getting closer to." Creates authenticity by mimicking doc-style camera work. Shows the subject from multiple angles, communicating confidence.
Traditional testimonials fail because they're static talking heads in flat lighting. Cinematic social proof makes viewers feel the transformation. Show the result first, the emotion second, the person third. The first two seconds determine whether the viewer stays. Every testimonial video needs an i
1. Never show a flat screenshot. Always add depth — perspective, reflection, environment, motion. 2. The UI is the hero, not the narrator. Show the interface doing things, not someone explaining it. 3. Sound sells more than visuals. A satisfying click, a smooth transition sound, a bass hit on featur
A complete system for crafting Seedance 2.0 prompts on Higgsfield that feature AI avatars, digital personas, virtual presenters, and synthetic characters as the primary subject. This skill covers everything from photorealistic digital humans to stylized 3D characters, including hook patterns, enviro
When generating a prompt, collect or infer: Seedance 2.0 prompt structure: Each prompt block should be 15–25 lines covering: scene, camera, lighting, timing breakdown, motion notes, and sound direction.
Turn social media paper recommendations into actionable research items. Use platform-specific tools to fetch the full content: From the extracted content, identify all referenced papers:
Systematic literature survey for positioning a research contribution. Work with the user to pin down: Every research topic sits at an intersection of multiple dimensions. Identify 2-4 axes:
Structured workflow for reading academic papers efficiently. Follow the appropriate level template above. When in doubt, start with Level 2. After reading, save the digest:
Pipeline for downloading academic paper full-text at scale. Handles the three classes of sources that exist in 2026: 1. Publisher TDM APIs (Elsevier / Wiley / Springer) — for paywalled content where the institution has a subscription 2. OA aggregators (Unpaywall / OpenAlex / Crossref) — for Open Acc
Systematic, multi-engine academic paper search. Choose engines based on the search goal: Before searching, clarify:
Systematic management of an academic literature library via Zotero. Standard collection hierarchy: Use tags to cross-cut the collection hierarchy:
Thin CLI wrapper around PaperBanana (a.k.a. PaperVizAgent), a multi-agent figure-generation pipeline for academic papers. The skill provides exactly one script: . It feeds
