cangjie-skill

把书、长视频、播客等高价值内容蒸馏成可执行的 Agent Skills

133 articles 3.5k View on GitHub ↗
133 articles
How Atomic Skill Design Prevents Anti-Patterns in cangjie-skill

Discover how atomic skill design in cangjie-skill prevents anti-patterns. Learn about the six invariants that ensure robust, maintainable, and verifiable skill development.

best-practices
Aug 16, 2026
How to Apply RIA拆书法 Principles Beyond Books: A Complete Guide to Media-Agnostic Skill Extraction

Apply RIA拆书法 principles beyond books to videos podcasts and courses Extract skills from any media using the Cangjie Skill pipeline with transcripts

how-to-guide
Aug 16, 2026
How User Confirmation Checkpoints Prevent Rework in Stages 0 and 1.5 of the RIA-TV++ Pipeline

User confirmation checkpoints in Stages 0 and 1.5 of the RIA-TV++ pipeline validate skeleton structure and candidate lists early, preventing costly rework.

how-to-guide
Aug 16, 2026
How to Install Generated Skills to Claude Code or Cursor: Copy-to-Skills Method

Learn the copy-to-skills method to install generated cangjie-skill for Claude Code or Cursor. Manually add custom skills to your IDE to activate them.

how-to-guide
Aug 16, 2026
How the DIGEST.md Format Differs from Simple Summarization: A Structured Knowledge Distillation System

Discover how DIGEST.md's structured knowledge distillation surpasses simple summarization with triple-verified extracts, explicit failure modes, and navigable links.

deep-dive
Aug 16, 2026
How GLOSSARY.md Functions as a Shared Terminology Dictionary in cangjie-skill

Discover how GLOSSARY.md acts as a shared terminology dictionary in cangjie-skill, ensuring consistent vocabulary and preventing misinterpretations across AI skills.

internals
Aug 16, 2026
How Cross-Skill Confusion Testing Works with Sibling Skills in Cangjie-Skill

Learn how cross-skill confusion testing in Cangjie-Skill ensures sibling skills don't interfere. Discover unique skill identification with ambiguous prompts for robust LLM execution.

how-to-guide
Aug 16, 2026
How the INDEX.md with Mermaid Citation Graphs Is Generated in cangjie-skill

Discover how cangjie-skill automatically generates INDEX.md with Mermaid citation graphs using Jinja2 templates and Zettelkasten linking to visualize skill relationships like depends-on, contrasts-with, and composes-with.

internals
Aug 16, 2026
How to Write a `test-prompts.json` File in Darwin-Compatible Format for Cangjie Skill

Learn to write test-prompts.json in Darwin-compatible format for Cangjie skill. Ensure reliable macOS parsing with UTF-8 encoding, LF line endings, and no trailing commas.

how-to-guide
Aug 16, 2026
How the Cangjie-Skill Checkpoint and Resume Mechanism Uses PIPELINE_STATE.md

Discover how the Cangjie-Skill checkpoint resume mechanism leverages PIPELINE_STATE.md to reliably restart interrupted pipelines. Learn to track progress with a simple markdown checklist.

internals
Aug 16, 2026
How Stage 0 Adler Analysis Extracts Book Structure and Critique in Cangjie‑Skill

Discover how Stage 0 Adler analysis extracts book structure and critique using four systematic steps, generating a ready-to-use BOOK_OVERVIEWmd artifact for further skill extraction.

deep-dive
Aug 16, 2026
How to Define Accurate Trigger Conditions (A2 Field) for Callable Skills

Master defining accurate trigger conditions for callable skills in the A2 field. Ensure your skill activates correctly with this essential guide.

how-to-guide
Aug 16, 2026
…

Have a question about this repo?

These articles cover the highlights, but your codebase questions are specific. Give your agent direct access to the source. Share this with your agent to get started:

Share the following with your agent to get started:
curl -s "https://instagit.com/install.md"

Works with
Claude Codex Cursor VS Code OpenClaw Any MCP Client

Maintain an open-source project? Get it listed too →