skill-contract-reviewer 精选
Validate an Agent Skill package and choose the right instruction, capability, or lifecycle primitive before implementation.
安装方式
cp -r phases/13-tools-and-protocols/22-skills-and-agent-sdks/outputs/skill-contract-reviewer ~/.claude/skills/skill-contract-reviewer把该 skill 目录放进客户端 skills 目录。不同客户端的目录可能不同,Claude Code 默认是 ~/.claude/skills/
https://raw.githubusercontent.com/rohitg00/ai-engineering-from-scratch/56c886c8821fc488b6f9608895c983d18f0d1ea3/phases/13-tools-and-protocols/22-skills-and-agent-sdks/outputs/skill-contract-reviewer/SKILL.md下载后另存为 ~/.claude/skills/skill-contract-reviewer/SKILL.md。链接锁定采集时的 commit,内容与本站展示一致。
整仓一次性获取:git clone --depth 1 https://github.com/rohitg00/ai-engineering-from-scratch.git
依赖与权限
- 无需 API Key
- 无需联网
- 依赖外部命令:
python3 - 以上为按正文与 frontmatter 自动推断的信号,仅供参考;安装前建议先阅读原文。
SKILL.md 正文
Skill contract reviewer
Use this skill when a workflow is about to become a reusable agent artifact.
- Set
SKILL_ROOTto the absolute directory containing this installed
SKILL.md. Do not assume the process working directory is the bundle.
- Set
TARGET_ROOTto the absolute original workspace working directory and
resolve the proposed skill directory under that root.
- Read
$SKILL_ROOT/references/contract.mdand validate the portable
SKILL.md identity fields.
- Read
$SKILL_ROOT/references/decision-model.mdand separate repository
context, reusable method, external capability, lifecycle timing,
deterministic logic, and isolated delegation.
- Before execution, show the exact resolved argument vector. Run
python3 "$SKILL_ROOT/scripts/check_skill.py" "$TARGET_SKILL", where
TARGET_SKILL is the absolute proposed skill directory under
TARGET_ROOT.
- Inspect the JSON report. Fix every error before discussing host-specific
extensions.
- Compare the proposed artifact with
$SKILL_ROOT/assets/task-shapes.json and return the smallest composable set
of primitives.
Do not claim that a runtime extension is part of the portable contract. Do not treat a valid skill as permission to run scripts or access tools.
Return the validation report, the selected primitives, and one sentence
explaining each selection. Include execution evidence with the resolved script
path, resolved target path, cwd, exact argv, and exit code. If the host cannot
expose one of those observations, mark it unverified instead of inventing it.
质量与来源
- 所属仓库
- rohitg00/ai-engineering-from-scratch
- 仓库星标
- 6.8 万
- 最近更新
- 昨天
- 采集时间
- 今天
- commit
- 56c886c8821f
- 内容指纹
- sha256:84345705cb4959f
- 文件路径
- phases/13-tools-and-protocols/22-skills-and-agent-sdks/outputs/skill-contract-reviewer/SKILL.md