An advanced knowledge-distillation meta-skill that transforms methodologies inside books into atomic, executable AI agent skills and human-readable learning assets.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install cangjie-skill
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install cangjie-skill using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is book2skill?
book2skill is a highly structured meta-skill designed to process deep knowledge from text sources (PDF, EPUB, TXT) and refine them into executable formats for AI agents. Rather than generating simple book summaries or character roles, it systematically extracts decision frameworks, core principles, and actionable checklists. It optimizes these extractions so they can be natively executed by automated coding agents.
By converting raw reading material into the structured Openclaw Skills format, this tool bridges the gap between static human literature and dynamic AI execution. Version 2.0.0 introduces an advanced humanized output layer, producing conversational learning notes and real-world talking points alongside rigorous automated testing configurations to guarantee production-ready agent performance.
book2skill Use Cases
- AI Skill Engineering: Transforming an industry-standard framework or methodology textbook into interactive, executable tasks for downstream AI agents.
- Knowledge Asset Distillation: Creating clean, conversational learning notebooks and real-world communication templates from dense academic or professional books.
- Agent Behavior Standardization: Hardening soft concepts from business, management, or technical books into structured code environments through precise test prompt synthesis.
How book2skill Works
- Phase 0 (Adler Understanding): Reads the input text and applies a four-step comprehensive structure, interpretation, critique, and application analysis to map out the foundational blueprint into a book overview file.
- Phase 1 (Parallel Extraction): Spawns five concurrent task sub-agents specialized in framework, principle, case study, counter-example, and glossary extraction to create a raw candidate pool.
- Phase 1.5 (Triple Verification): Filters each candidate framework across cross-domain validity, predictive power, and non-trivial uniqueness metrics, offloading failures to an audit directory.
- Phase 2 (RIA++ Construction): Synthesizes valid items into individual skill files containing precise reading references, custom interpretations, historical test cases, future trigger parameters, execution steps, and operational boundaries.
- Phase 3 (Zettelkasten Linking): Identifies dependencies or contrasts between isolated modules and builds an index map enhanced with relational visual diagrams.
- Phase 4 (Pressure Testing): Automatically designs and localizes 5 to 10 test prompts (including explicit honeypots and ambiguous scenarios) to measure compatibility with evolutionary tools.
- Phase 5 (Human Output Creation): Generates lightweight learning summaries and deployment conversation scripts designed to read naturally like casual human-to-human exchanges.
book2skill Setup
Ensure you have your book text source ready in an accessible plain text or document file path before invoking the skill. book2skill runs directly within Openclaw environments.
# Invoke the meta-skill directly via the CLI runner
openclaw run book2skill --book "/path/to/methodology_book.pdf"
book2skill Data Schema & Taxonomy
The skill generates a clean taxonomy under a targeted books/<book-slug>/ repository branch to preserve strict audit trails:
| File / Folder Path | Stage | Description |
|---|---|---|
BOOK_OVERVIEW.md |
Phase 0 | Structural blueprint, overarching themes, core terminology, and technical critiques. |
INDEX.md |
Phase 3 | Central skill registry including an integrated Mermaid diagram mapping references. |
LEARNING_NOTE.md |
Phase 5 | A conversational, human-friendly breakdown containing analogies and simplified explanations. |
TALKING_POINTS.md |
Phase 5 | A library of at least 10 real-world verbal scripts categorized by contextual situation tags. |
candidates/ |
Phase 1 | Contains the raw extraction exports produced by the parallel sub-agents. |
rejected/ |
Phase 1.5 | Catalog of components that failed triple verification, complete with explicit reasons. |
<skill-slug>/SKILL.md |
Phase 2 | The production-ready markdown configuration for individual Openclaw Skills. |
<skill-slug>/test-prompts.json |
Phase 4 | Hardened functional prompt test suites built to comply with darwin-skill evolution formats. |
book2skill Advanced Features
- Parallel Sub-Agent Pipelines: Leverages tool-driven concurrent operations to process text across multiple analytical viewpoints in parallel, optimizing resource usage.
- Cost-Optimized Dual Model Routing: Recommends lightweight, cost-effective inference engines like doubao-lite or DSF for intensive parallel tasks to reduce latency and execution overhead.
- Ecosystem Interoperability: Generates localized validation schemas configured natively for seamless integration with darwin-skill auto-evolution routines.
- Strict Error Handling & Rollbacks: Enforces automated recovery protocols that trigger complete generation rollbacks if test suite passing rates drop below an 80% baseline threshold.
SKILL.md
Loading
Related Openclaw Skills
E-commerce Image & Detail Page Diagnosis Assistant
An AI-powered diagnostic agent that analyzes e-commerce product images, search visibility, and detail pages to generate comprehensive, visually structured HTML optimization reports.
EPUB Bilingual Converter Skill
A comprehensive pipeline for translating and generating paragraph-aligned, bilingual EPUB books and newsletters with an AI-guided workflow.
Histrategy
Histrategy is an AI-powered historical strategy game that allows players to command armies using natural language in scenarios like the Three Kingdoms or Rome Triumvirate.
Skill Selector
Skill Selector is an orchestration utility that analyzes user intent and dynamically targets the most suitable installed capability using semantic and keyword matching.
VueUse Functions Skill
A specialized AI skill designed to automatically map requirements and apply optimal VueUse composables in Vue.js and Nuxt projects.
Web Design Guidelines
An automated UI and accessibility review tool that audits your web interface code against live, fetched design guidelines.
Featured*
