book-to-skill

book-to-skill is an open-source tool that converts user-owned documents and document folders—such as technical book PDFs, EPUBs, DOCXs, HTML, Markdown, and RTFs—into structured skills that AI coding agents can utilize. Instead of simply putting the entire document into the conversation context, a Python-based extractor cleans the text and metadata from the source document, and the agent uses it to generate core concepts, chapter-by-chapter files, glossaries, patterns, and cheat sheets. The generated skills load only the necessary chapters on demand, reducing the cost and token usage of repeatedly reading long documents. It can be used in environments supporting the Agent Skills standard, such as GitHub Copilot CLI, Amp, and Claude Code, and processes files locally without uploading them to external services. It primarily targets developers and technical teams who want to study technical books, search internal documentation, reference architectures and runbooks, organize research materials, and utilize standards and specification documents. It is distributed under the MIT license and provides a command-line interface and optional document extractor dependencies.