Capability Distill

Capability Distill is an AI Engineering skill for Claude Code, published by majiayu000 in spellbook.

265 stars26 forkson majiayu000/spellbookAdded 2026/09/02+1% in starsRepository updated 2026/08/31

agent-skillsai-agent-skillsai-agentsai-coding-assistantautomationclaudeclaude-codeclaude-code-skillscode-reviewcodexcodex-skillscross-runtimedeveloper-toolsmulti-agentproductivityprompt-engineeringskill-librarysoftware-developmentspellbookworkflows

Install in seconds

Install Capability Distill

Copy Capability Distill into your Claude Code skills folder. Run the command in your terminal, or review the source on GitHub before installing.

terminal

npx degit https://github.com/majiayu000/spellbook/tree/main/skills/capability-distill ~/.claude/skills/capability-distill

Requires Node.js. Downloads this skill only — not the rest of the repository — into your Claude Code skills folder.

Without Node.js

git clone https://github.com/majiayu000/spellbook.git

Clones the whole repository, then copy the skill’s own directory into your skills folder yourself.

In this catalog

Source file: skills/capability-distill/SKILL.md in majiayu000/spellbook
Installs to: ~/.claude/skills/capability-distill
Collection: One of 25 skills cataloged from this repository
Category: AI Engineering — 2631 skills

What Capability Distill does

Capability Distill extracts non-obvious judgment rules from approved strong-model interviews or real task traces into an auditable judgment packet. Use it when you want to preserve key decisions, not when you are just writing process docs or a SKILL.md from scratch.

Capability Distill is cataloged under AI Engineering on DirSkills. Capability Distill comes from a repository tagged agent-skills, ai-agent-skills, ai-agents, ai-coding-assistant and automation.

Documentation

README

Excerpt

Capability Distill

把“为什么在两个合理选项中选择其中一个”提取成证据支持的判断规则。不要把操作步骤、通用建议或原始会话内容换个格式包装成 skill。

与现有技能的边界

capability-distill：选择获准证据、还原决策事件、提取判断规则、输出 judgment packet。
skill-audit：判断这些规则是否值得成为 skill、归属哪个现有 skill、是否重复以及如何分层。
skill-creator：创建或修改具体的 SKILL.md、设计 with-skill/baseline eval、迭代和验证触发描述。
skill-lifeguard：当产物属于高影响工作流时补可靠性契约和漂移修复闭环。

不要在本 skill 中复制后三者的完整写作、注册或评测流程。目标是提供它们可消费的高信号输入。

Operating Contract

Direct actions: 使用当前对话和用户明确批准的本地材料；先读元数据再读内容；生成脱敏的 judgment packet；运行本地只读检查。
Escalate before: 读取未获准的会话历史、memory、home 目录或其他仓库；把任何轨迹内容发送给外部模型；覆盖已有 skill；写入共享仓库；发布或安装产物。
Evidence-backed pushback: 判断密度低、证据不足或已有 skill 完整覆盖时，给出具体重叠项并建议不蒸馏或只更新原 skill。
Feedback loop: 把 eval 中的“未生效 / 机械化 / 误触发”连同对应 rule_id 回写到下一轮 judgment packet，而不是只改措辞。

0. 建立数据边界

在读取额外材料前记录以下字段：

This is the opening of the README. Read the full README on GitHub.

Frequently asked about Capability Distill

What else does majiayu000 publish alongside Capability Distill?
Capability Distill is one of 25 skills that DirSkills catalogs from majiayu000/spellbook, the repository it ships in. Its siblings there include AGENTS.md Scaffold, API Design and Agents Md Optimize. Each one is a separate skill with its own page in this directory, installs the same way Capability Distill does, and is maintained by majiayu000 in that same repository. The rest of the collection is listed on the majiayu000/spellbook page.
How does Capability Distill compare to other AI Engineering skills?
Capability Distill ranks #2232 by stars among the 2631 AI Engineering skills in this catalog. The most-starred ones next to it are Architecture Decision Records, AI-First Engineering and Agentic OS. DirSkills ranks by the star count of the repository each skill ships in, so that order reflects how popular those repositories are rather than any review of Capability Distill against them. Open each page to compare what they document and how they install.