AIdu Distill: Books, Courses, and Videos
Paste the following prompt into your AI chat to install this skill:
Please install @user_a58e0a33/aidu-distill using https://skillhub.cn/install/skillhub.md.
About this skill
The Problem
aidu-distill is not for book summaries or quote curation. It targets a specific gap: turning durable methodologies from long books, courses, and videos into skills that an agent can invoke in real tasks. It is especially useful for content that feels known but rarely applied, such as decision frameworks, principle lists, concept systems, and failure modes. A key constraint is that it does not distill from memory: if the source text is missing, it pauses and asks for material, reducing the risk of presenting model confabulation as the author’s idea.
How It Works
The pipeline has two modes. Lite suits single articles or short courses and produces 2–4 skills with a lighter flow. Standard suits whole books, video series, or podcast collections, runs the full seven-stage flow, and is better resumed across sessions. It starts with whole-text understanding, extracts candidate frameworks, verifies them across applicability, predictive power, and distinctiveness, then writes them into SKILL.md using R/I/A/E/B dimensions. There are three confirmation points: skeleton confirmation, selection confirmation, and budget confirmation. At delivery, it presents a skill list first and installs only the skills the user selects, preventing the user-level skill directory from becoming cluttered.
Boundaries
It works best on content with high methodology density. Essays, quote collections, and persona-roleplay material are poor fits. Video and podcast input should first become transcripts. Every skill must be traceable, have clear trigger conditions, and pass quality checks; if the source text is vague or the user does not confirm priorities, output quality drops noticeably.
Use Cases
- Turn a product decision book's judgment frameworks into five callable skills for later review tasks.
- Extract recurring review principles from course transcripts into agent skills with clear trigger conditions.
- Use Lite mode on a podcast transcript to distill one episode into three previewed skills.
- Filter failure modes from a book into reusable checks while recording why weak candidates were rejected.
Best For
- Product leads who need to turn long courses into reusable frameworks for review meetings.
- Engineers maintaining personal knowledge bases and want book principles as callable agent skills.
- Knowledge operators processing podcast or video transcripts to extract methodology, not summaries.
- Researchers sharing team decision checklists without installing every generated skill.
Related Skills
Updates Next.js documentation based on code changes in the active branch to help maintainers review PR documentation completeness.
Use a book title and author to search reviews, contents, and reader feedback, then generate a structured breakdown of historical or decision-logic books with key claims, concepts, quotes, and reading notes in Obsidian.
A multi-system divination advisor covering Bazi, Ziwei, tarot, feng shui, timing, and naming with structured interpretations.
Provides Bazi, Liu Yao, Qimen, Meihua, and date-selection guidance from classical references, with plain-language explanations and actionable suggestions.