honnibal/claude-skills

Registry 信息

项目简介

由 honnibal 创建的个人实验性仓库，用于探索 Claude Skill，使用前请仔细审阅。

要点

• 实验性 Claude Skill 集合
• 基于 Python 的实现
• 使用前需谨慎审阅
• 个人探索项目

使用场景

• 测试新的 Claude Skill 想法
• 学习如何构建 Claude Skill
• 原型设计 AI 助手功能

README 摘要

# claude-skills Reusable [Claude Code](https://docs.anthropic.com/en/docs/claude-code) slash command skills. To install a skill, copy it to `~/.claude/commands/` (rename from `.md.txt` to `.md`). The skills and parts of the project documentation were produced semi-automatically using Claude Code. You can read more about the general thinking behind these skills on [my blog](https://honnibal.dev). ## Skills All skills accept an optional argument to scope the work to specific files or directories. If no argument is given, they work through the project or ask you to choose a starting scope. The skills are set to `disable-model-invocation: true` by default. This means installing them shouldn't pollute your context. You can change this if you want the model to invoke them automatically. ### `tighten-types.md.txt` Systematically review Python source files and tighten type annotations. Works through a prioritised checklist: missing class attribute annotations, vague `Any` types replaced with concrete library types, structured dicts promoted to Pydantic models or `TypedDict`, `@overload` signatures for narrowable unions, redundant in-body annotations fixed at the root cause, and style modernisation (`Optional[X]` → `X | None`, etc.). ``` /tighten-types src/mypackage/core.py ``` ### `contract-docstrings.md.txt` Write docstrings that document each function's **contract** --- what it requires of callers, what it guarantees, and how it fails. Analyses four dimensions: input invariants (preconditions beyond the type signature), errors raised on violation (explicit checks vs. implicit crashes vs. silent wrong results), errors from external state (filesystem, network, databases), and silenced errors (broad `except` blocks, `.get()` defaults, `suppress()`). Produces a structur

话题

暂无话题