القائمة
Use when verifying that one broken skill does not abort the corpus run.
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
استنادا إلى تصنيف SOC المهني
I help with tidying up messy notes whenever it seems useful.
Use when the user asks to create, write, compose, or design a new agent skill — turning an idea, notes, requirement, or repeated workflow into a SKILL.md with eval cases and a trigger set through an interview → draft → critique → refine loop.
Use when the user asks to lint, audit, test, benchmark, validate, or fix an agent skill — from a single SKILL.md frontmatter check to trigger-accuracy measurement or a corpus-wide audit of installed skills for duplication — driving the shakespii CLI (init, lint --json, test --run, bench) to resolve findings until clean.
A tiny practice skill that summarizes shell command output into one line.
Use when the user asks for a one-word acknowledgement drill — replies with exactly the word the drill names.
Compress natural language memory files (CLAUDE.md, todos, preferences) into caveman format to save input tokens. Preserves all technical substance, code, URLs, and structure. Compressed version overwrites the original file. Human-readable backup saved as FILE.original.md. Trigger: /caveman:compress <filepath> or "compress memory file"
| name | corpus-good |
| description | Use when verifying that one broken skill does not abort the corpus run. |
| version | 0.1.0 |
Serves as the healthy neighbor of a deliberately broken fixture skill.
The corpus root path handed over by the broken-skill test.
The broken fixture tree is checked out unchanged.
An empty findings list for the good fixture.
Given the input good, the expected output is good-lint-clean.
Fixing the broken neighbor; it is broken by design.
