AI Agent Skills

针对高可信度的一手来源调查一个问题，并把发现记录为仓库中的一个 Markdown 文件。当用户想要调研某个主题、收集文档或 API 事实，或把阅读跑腿活交给后台 agent 时使用。

Working with Microsoft Access databases exported via VCS (Version Control System for Access). Use when editing Access form definitions (.bas), report definitions, query…

Turn one URL into clean markdown for a model to read. Use when a user gives you a link, when you need a page as grounding, or when a previous fetch returned a JavaScript…

Design human-visible, machine-readable website layers for AI assistants, RAG, and agentic search. Use this skill when the user asks for llms.txt, LLM discovery plans, vi…

Framework for deep research reports: define the question and tier sources first, cross-verify, then write conclusion-first reports with explicit uncertainty statements.…

Proofreads English technical writing for typos, grammar slips, punctuation, terminology consistency, jargon, and structure; returns an itemized Original → Suggestion → R…

Turn abstract concepts in a user-provided book PDF, paper, textbook, lecture note, or long document into source-grounded stories, visual memory anchors, real-life signal…

Interviews someone about their life — what they know, who they understand, what they have available — then shapes 3–4 business ideas built from their answers and finishe…

Turn a verified BaZi-derived life-cycle data series into a traceable Life K-Line report for stage planning. Use for 人生K线, 人生曲线, 阶段高低点, 机会窗口, or long-range review; not fo…

Use thirds.ai to make a one-off PDF or image with MCP, or find the right skill for a brand kit, reusable template, or batch.

Make one PDF or image per row from a saved thirds.ai design, then check row results and collect the archive through MCP.

Make an editable thirds.ai PDF or image template from a gallery design, HTML, words, or a source image, then review it before saving.

Fill a saved thirds.ai template with new data and make a branded PDF, PNG, JPEG, or WebP through MCP.