Repository navigation
Codebase to Mastery v1.0.0
Codebase to Mastery v1.0.0
首个正式版本 / First stable release.
Turn a repository into a progressive, source-grounded learning kit: orient, trace, reason, rebuild, extend and teach back. Includes an offline interactive course with the refreshed dark chapter directory, teal accents and mobile chapter strip.
将真实代码库转化为由浅入深的学习包:认识全貌、追踪源码、解释设计、局部重建、扩展验证、复述迁移。包含精确源码引用、主动回忆、starter/reference 练习、学习记录和增量更新。
Install / 安装
Install this exact release in your project:
npx skills add https://github.com/StormTian/codebase-to-mastery/tree/v1.0.0 \
--skill codebase-to-mastery --agent codex claude-codeAdd --global for a personal installation. This uses the open-source Skills CLI; this release is an Agent Skill distribution, not an independently published npm executable.
Or download the ZIP / tar.gz below. Both extract to codebase-to-mastery/. Place that complete folder under .agents/skills/ for a Codex project or .claude/skills/ for a Claude Code project. For personal installation, use your client's documented skill directory. Requires Python 3.10+ and Git for the helper workflow; generated courses work offline.
也可下载下方 ZIP 或 tar.gz,解压后将完整的 codebase-to-mastery/ 放到客户端的 Skill 目录。保留脚本、模板和参考资料的相对路径。
Assets and verification / 发布文件与验证
codebase-to-mastery-v1.0.0.zip: complete installable Skill folder.codebase-to-mastery-v1.0.0.tar.gz: equivalent payload for tar-based workflows.release-manifest.json: exact source commit, per-file SHA-256 and archive checksums.SHA256SUMS: checksums for both archives and the manifest.
Download all four files into one directory, then verify:
sha256sum --check SHA256SUMS
# macOS alternative:
shasum -a 256 -c SHA256SUMSThe release workflow runs Python 3.10/3.12 tooling checks and original-demo controls before publishing, then exercises the extracted ZIP. Release packaging excludes repository-maintenance configuration and reads only the tagged commit. Local learner answers and machine-specific run receipts are not part of the package.
Installation compatibility and passing original fixtures do not establish learner mastery or production Agent behavior. See validation boundaries. MIT license for original repository material; other projects' teaching excerpts retain their own terms.