Back to Details

skill-creator

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

Skill Creator

Skill Creator

Research-enhanced skill creator that produces higher-quality skills than built-in defaults. Every design decision is grounded in how transformers process context.
研究增强型Skill创建器，生成的Skill质量优于内置默认方案，所有设计决策都基于Transformer处理上下文的底层逻辑。

Expert Vocabulary Payload

专家词汇载荷

Prompt Engineering & Routing: expert vocabulary payload, dual-register description, vocabulary routing, embedding space routing, attention budget, distribution center, right altitude, retrieval anchor
Skill Architecture: progressive disclosure, context window management, U-shaped attention curve, YAML frontmatter, trigger surface, structural delineation, three-level loading (metadata / SKILL.md / references)
Behavioral Design: anti-pattern watchlist, detection signal, counter-example, imperative instruction, conditional branching, evaluation criteria
Quality & Testing: canonical example, few-shot learning, 15-year practitioner test, consultant-speak (banned), over-prompting, recency bias
提示词工程与路由： expert vocabulary payload, dual-register description, vocabulary routing, embedding space routing, attention budget, distribution center, right altitude, retrieval anchor
Skill架构： progressive disclosure, context window management, U-shaped attention curve, YAML frontmatter, trigger surface, structural delineation, three-level loading (metadata / SKILL.md / references)
行为设计： anti-pattern watchlist, detection signal, counter-example, imperative instruction, conditional branching, evaluation criteria
质量与测试： canonical example, few-shot learning, 15-year practitioner test, consultant-speak (banned), over-prompting, recency bias

Anti-Pattern Watchlist

反模式观察列表

These are anti-patterns in the skills this creator generates. Scan every generated SKILL.md against this list before delivery.
以下是本创建器生成的Skill中需要避免的反模式，交付前请对照此列表扫描所有生成的SKILL.md文件。

1. Generic Consultant-Speak in Vocabulary

1. 词汇中使用通用咨询套话

Detection: Vocabulary payload contains terms like "best practices," "leverage," "synergy," "robust solution," "scalable framework," or "holistic approach." Apply the 15-year practitioner test: would a senior domain expert use this exact term with a peer? If not, it fails. Resolution: Replace every generic term with the precise domain term it vaguely gestures at. "Best practices for error handling" becomes "circuit breaker pattern (Nygard), exponential backoff, dead letter queue."
检测方式： 词汇载荷包含「最佳实践」「 leverage」「协同」「健壮解决方案」「可扩展框架」「整体方法」等术语。应用15年从业者测试：资深领域专家会和同行使用完全相同的表述吗？如果不会，就不符合要求。 解决方案： 将每个通用术语替换为它所模糊指代的精确领域术语，比如「错误处理最佳实践」替换为「circuit breaker pattern (Nygard)、指数退避、死信队列」。

2. Over-Prompting

2. 过度提示

Detection: SKILL.md exceeds 500 lines. The same concept is stated 2-3 times in different words "for emphasis." Instructions contain hedging phrases ("you might want to consider," "it could be helpful to"). Resolution: State each instruction once, in imperative form. Remove hedging. Move heavy reference content to 
references/
. Test with a minimal version first; add detail only where the model demonstrably fails.
检测方式： SKILL.md超过500行，同一个概念用2-3种不同表述「强调」，指令包含模糊 hedging 表述（「你可能需要考虑」「这可能会有帮助」）。 解决方案： 每条指令用祈使语气仅表述一次，删除模糊表述，将 heavy reference 内容移动到
references/
目录下。先使用最小版本测试，仅在模型明确表现出能力不足时再补充细节。

3. Positive-Only Instructions

3. 仅包含正向指令

Detection: Zero "do NOT" or "avoid" guidance. No anti-pattern watchlist. The skill only describes what to do, never what not to do. Resolution: Add 5-10 domain-specific anti-patterns with named patterns, detection signals, and resolution steps. Without negative constraints, the model gravitates to the distribution center (the most generic, average output).
检测方式： 没有「禁止」或「避免」类指引，没有反模式观察列表，Skill仅描述要做什么，完全不提及不能做什么。 解决方案： 添加5-10个领域专属反模式，包含命名模式、检测信号和解决步骤。没有负向约束的情况下，模型会倾向于生成分布中心内容（最通用的平均输出）。

4. Single-Register Description

4. 单寄存器描述

Detection: Description uses only formal terminology OR only casual language. Test: would the skill trigger if a user said "help me with [casual version of task]"? If not, the casual register is missing. Resolution: Rewrite description to include both expert terms (for routing to deep knowledge) and natural-language trigger scenarios (for reliable activation). Add explicit "even if they don't say [formal term]" phrases.
检测方式： 描述仅使用正式术语或者仅使用 casual 口语。测试：如果用户说「帮我处理[任务的口语化表述]」，该Skill会触发吗？如果不会，说明缺少口语化寄存器。 解决方案： 重写描述，同时包含专家术语（用于路由到深度知识）和自然语言触发场景（保障可靠激活），添加明确的「即使用户没有提到[正式术语]」类表述。

5. Edge-Case Stuffing

5. 堆砌边缘场景规则

Detection: More than 15 specific edge-case rules. Long lists of "if X then Y" covering every scenario instead of demonstrating the pattern. Resolution: Replace with 2-3 diverse canonical examples that show the pattern. Include one hard case. Let the model generalize from examples rather than memorize rules. Research shows 2-3 examples often match the effectiveness of 9+.
检测方式： 有超过15条特定边缘场景规则，用长长的「如果X则Y」列表覆盖所有场景，而不是展示底层模式。 解决方案： 替换为2-3个覆盖不同场景的规范示例，包含1个复杂场景。让模型从示例中归纳规律，而非死记硬背规则，研究表明2-3个示例的效果通常和9个以上的示例相当。

6. Paragraph-Form Logic

6. 段落形式逻辑

Detection: Complex multi-step behavior described in prose paragraphs. No numbered steps, no IF/THEN conditions, no imperative verbs. Resolution: Refactor to imperative ordered steps with explicit conditions. "First check for anti-patterns, and if you find some you should probably address them" becomes "1. Scan input for anti-patterns. IF detected: apply Detect-Name-Explain-Resolve-Prevent. IF none: proceed to step 2."
检测方式： 复杂多步骤行为用散文段落描述，没有编号步骤，没有IF/THEN条件，没有祈使动词。 解决方案： 重构为带有明确条件的祈使式编号步骤，「首先检查反模式，如果发现的话你可能需要处理它们」改写为「1. 扫描输入中的反模式，IF 检测到：应用检测-命名-解释-解决-预防流程；IF 未检测到：进入步骤2。」

7. Missing Examples

7. 缺少示例

Detection: Zero input-to-output examples or BAD/GOOD pairs in the generated skill. The skill relies entirely on verbal instructions. Resolution: Add 2-3 diverse examples. Use BAD vs GOOD pairs for quality standards, input-to-output pairs for workflows. Place the most representative example last (recency bias gives it the strongest influence).
检测方式： 生成的Skill中没有输入输出示例或者错误/正确对比对，完全依赖文字指令。 解决方案： 添加2-3个覆盖不同场景的示例，用错误vs正确对体现质量标准，用输入输出对体现工作流。将最具代表性的示例放在最后（近因偏差会让它对输出的影响最强）。

Behavioral Instructions

行为指令

Phase 1: Capture Intent

阶段1：捕获意图

  1. Identify what the skill should do, when it should trigger, and what output format is expected.
  2. Identify the target audience: who triggers this skill, and what do they typically say?
  3. IF the user's request is vague: ask 1-2 clarifying questions about domain, trigger scenarios, and expected output format. Do NOT proceed with a generic skill.
  1. 明确Skill需要实现的功能、触发场景以及预期输出格式。
  2. 明确目标受众：谁会触发这个Skill，他们通常会怎么表述需求？
  3. IF 用户需求模糊：询问1-2个关于领域、触发场景、预期输出格式的澄清问题，不要生成通用Skill。

Phase 2: Research (Conditional)

阶段2：调研（条件执行）

  1. IF unfamiliar domain: research the domain before drafting.
    • Use web search for domain terminology, frameworks, and named patterns.
    • Use codebase exploration if the skill is project-specific.
    • Identify 15-30 expert terms that pass the 15-year practitioner test.
    • Identify 5-10 domain anti-patterns with established names.
  1. IF 不熟悉目标领域：在起草前先调研相关领域。
    • 用网页搜索获取领域术语、框架和命名模式。
    • 如果是项目专属Skill，探索代码库。
    • 筛选15-30个通过15年从业者测试的专家术语。
    • 筛选5-10个有公认命名的领域反模式。

Phase 3: Draft SKILL.md

阶段3：起草SKILL.md

  1. Write YAML frontmatter with dual-register description (~100 words, pushy).
    • Include both expert terminology AND natural-language trigger scenarios.
    • Include explicit exclusions ("Do NOT use for...").
    • Be aggressive about triggering — current models undertrigger. Include synonyms, edge cases, and adjacent scenarios.
  2. Write Expert Vocabulary Payload (15-30 terms in 3-5 clusters).
    • Place BEFORE behavioral instructions. WHY: vocabulary primes the routing signal before execution begins.
    • Organize in sub-domain clusters of 3-8 terms each.
    • Include originator attribution for named frameworks: "circuit breaker (Nygard)" not just "circuit breaker."
    • Apply the 15-year practitioner test to every term. Remove any that fail.
  3. Write Anti-Pattern Watchlist (5-10 named patterns).
    • Place BEFORE behavioral instructions. WHY: the model checks for problems before proceeding with normal execution.
    • Each pattern needs: name + origin, detection signal, resolution step.
    • Use established names where they exist: "Bikeshedding (Parkinson)" not "spending too much time on details."
  4. Write Behavioral Instructions (imperative, ordered, conditional).
    • Use imperative verbs: "Scan," "Classify," "Output," not "You should try to."
    • Number every step. Use IF/THEN for branching logic.
    • Include WHY for non-obvious steps. WHY: the model can generalize principles to edge cases, but dead rules only cover literal matches.
    • Start with an anti-pattern scan step.
  5. Write Output Format specification.
    • Define required fields, structure, and templates.
    • Use structured formats (YAML, tables, numbered lists) over prose.
  6. Write 2-3 diverse Examples (BAD vs GOOD or input-to-output).
    • Cover different cases including at least one hard case.
    • Place the most representative example LAST. WHY: recency bias gives the final example the strongest influence on output.
  7. Write "Questions This Skill Answers" section (8-15 natural-language queries).
    • Place at the END of the file. WHY: retrieval anchors benefit from end-of-context attention (U-shaped curve).
    • Use the exact phrasing a user would type, including casual and imprecise versions.
    • These function as retrieval anchors, self-documentation, and test cases simultaneously.
  1. 编写带有双寄存器描述的YAML frontmatter（约100字，倾向于主动触发）。
    • 同时包含专家术语和自然语言触发场景。
    • 包含明确的排除范围：「请勿用于……」。
    • 触发范围设置得更激进——当前模型触发率普遍偏低，包含同义词、边缘场景和关联场景。
  2. 编写专家词汇载荷（分为3-5组，共15-30个术语）。
    • 放在行为指令之前。原因：词汇会在执行开始前预热路由信号。
    • 按子领域分组，每组3-8个术语。
    • 命名框架要标注发起人：比如「circuit breaker (Nygard)」而不是仅写「circuit breaker」。
    • 对每个术语应用15年从业者测试，删除不符合要求的术语。
  3. 编写反模式观察列表（5-10个命名模式）。
    • 放在行为指令之前。原因：模型在执行常规流程前会先检查问题。
    • 每个模式需要包含：名称+来源、检测信号、解决步骤。
    • 尽可能使用公认名称：比如「Bikeshedding (Parkinson)」而不是「在细节上浪费太多时间」。
  4. 编写行为指令（祈使式、编号、带条件）。
    • 使用祈使动词：「扫描」「分类」「输出」，不要用「你应该尝试」。
    • 给每个步骤编号，用IF/THEN处理分支逻辑。
    • 非直观步骤要补充原因。原因：模型可以将原则泛化到边缘场景，但死板的规则仅能覆盖字面匹配的场景。
    • 以反模式扫描步骤作为开头。
  5. 编写输出格式规范。
    • 定义必填字段、结构和模板。
    • 优先使用结构化格式（YAML、表格、编号列表）而非散文。
  6. 编写2-3个覆盖不同场景的示例（错误vs正确对或输入输出对）。
    • 覆盖不同场景，至少包含1个复杂场景。
    • 将最具代表性的示例放在最后。原因：近因偏差会让最终示例对输出的影响最强。
  7. 编写「本Skill回答的问题」章节（8-15个自然语言查询）。
    • 放在文件末尾。原因：U型注意力曲线会让上下文末尾的检索锚点效果更好。
    • 使用用户实际会输入的表述，包含口语化和不精确的版本。
    • 这些内容同时承担检索锚点、自文档和测试用例的作用。

Phase 4: Reference Files (Conditional)

阶段4：参考文件（条件执行）

  1. IF the skill requires heavy reference content (pattern libraries, extended examples, checklists, evaluation criteria):
    • Create files in 
      references/
      directory.
    • Keep each reference file under 300 lines.
    • Include clear guidance in SKILL.md on WHEN to read each reference file.
  2. Keep SKILL.md under 500 lines total. IF over 500: move content to references/.
  1. IF Skill需要大量参考内容（模式库、扩展示例、检查清单、评估标准）：
    • 在
      references/
      目录下创建对应文件。
    • 每个参考文件控制在300行以内。
    • 在SKILL.md中明确说明何时需要读取各个参考文件。
  2. SKILL.md总长度控制在500行以内，IF 超过500行：将内容移动到references/目录。

Phase 5: Library Metadata

阶段5：库元数据

  1. Include library metadata in YAML frontmatter: name, version, domain tags, and compatibility notes.
  1. 在YAML frontmatter中包含库元数据：名称、版本、领域标签、兼容性说明。

Phase 6: Validate and Package

阶段6：验证和打包

  1. Scan the generated SKILL.md against the Anti-Pattern Watchlist above. Fix any violations.
  2. IF subagents are available: test against 2-3 realistic prompts that a real user would type (casual, imprecise, no formal vocabulary).
  3. Package the skill: validate structure, zip, and use present_files for installation.
  4. Save to 
    library/skills/
    and update 
    index.json
    .
  5. Log creation to 
    usage-log.jsonl
    .
  1. 对照上文的反模式观察列表扫描生成的SKILL.md，修复所有违规问题。
  2. IF 有可用的子Agent：用2-3个真实用户会输入的实际提示（口语化、不精确、没有正式术语）测试。
  3. 打包Skill：验证结构、压缩、用present_files执行安装。
  4. 保存到
    library/skills/
    目录并更新
    index.json
    。
  5. 在
    usage-log.jsonl
    中记录创建日志。

Output Format

输出格式

The primary output is a complete skill directory:
skill-name/
  SKILL.md              # Core instructions (<500 lines)
  references/           # Optional: heavy reference content
    [topic].md          # Each file <300 lines
SKILL.md internal structure (in this order):
  1. YAML frontmatter (name, description, metadata)
  2. Expert Vocabulary Payload
  3. Anti-Pattern Watchlist
  4. Behavioral Instructions
  5. Output Format
  6. Examples (BAD vs GOOD)
  7. Questions This Skill Answers
主要输出是完整的Skill目录：
skill-name/
  SKILL.md              # Core instructions (<500 lines)
  references/           # Optional: heavy reference content
    [topic].md          # Each file <300 lines
SKILL.md内部结构（按以下顺序排列）：
  1. YAML frontmatter（名称、描述、元数据）
  2. 专家词汇载荷
  3. 反模式观察列表
  4. 行为指令
  5. 输出格式
  6. 示例（错误vs正确对）
  7. 本Skill回答的问题

Examples

示例

Example 1: BAD vs GOOD Description

示例1：错误vs正确描述

BAD:
description: "Helps with code review."
Single-register. Not pushy. No expert terms. No trigger scenarios. No exclusions. This skill will almost never fire, and when it does, it will produce generic output.
GOOD:
description: |
  Performs structured code review using cyclomatic complexity analysis,
  connascence taxonomy, and conventional comments (Slaughter). Use when
  the user asks to review code, check a PR, look at their changes, or
  says "is this good?" about code -- even if they don't mention "review."
  Also triggers for diff review, merge request feedback, and pre-commit
  quality checks. Do NOT use for architecture decisions (use Decision
  Advisor) or writing new code (use Code Generator).
Dual-register. Pushy. Expert terms route to deep knowledge. Casual triggers ensure activation. Explicit exclusions prevent mis-triggers.
错误：
description: "Helps with code review."
单寄存器、不主动、没有专家术语、没有触发场景、没有排除范围。这个Skill几乎不会触发，就算触发也会生成通用输出。
正确：
description: |
  Performs structured code review using cyclomatic complexity analysis,
  connascence taxonomy, and conventional comments (Slaughter). Use when
  the user asks to review code, check a PR, look at their changes, or
  says "is this good?" about code -- even if they don't mention "review."
  Also triggers for diff review, merge request feedback, and pre-commit
  quality checks. Do NOT use for architecture decisions (use Decision
  Advisor) or writing new code (use Code Generator).
双寄存器、主动触发、专家术语路由到深度知识、口语化触发保障激活、明确的排除范围避免误触发。

Example 2: BAD vs GOOD Vocabulary Payload

示例2：错误vs正确词汇载荷

BAD:
undefined
错误：
undefined

Domain Vocabulary

Domain Vocabulary

good code, clean code, readable, maintainable, well-tested
Generic terms that every blog post uses. Routes to introductory content. Fails the 15-year practitioner test: no senior engineer says "good code" to a peer.

**GOOD:**
good code, clean code, readable, maintainable, well-tested
所有博客都在使用的通用术语，会路由到入门级内容，未通过15年从业者测试：没有资深工程师会和同行说「good code」。

**正确：**

Domain Vocabulary

Domain Vocabulary

Structural Analysis: cyclomatic complexity (McCabe), cognitive complexity (SonarSource), afferent/efferent coupling (Martin), connascence (Page-Jones)
Change Safety: shotgun surgery (Fowler), feature envy, divergent change, Liskov substitution violation
Review Process: conventional comments (Slaughter), ship/no-ship framework, diff review vs design review, LGTM criteria
Precise terms organized in clusters. Named frameworks with originators. Routes to code review expertise, not generic advice. Every term passes the 15-year practitioner test.

---
Structural Analysis: cyclomatic complexity (McCabe), cognitive complexity (SonarSource), afferent/efferent coupling (Martin), connascence (Page-Jones)
Change Safety: shotgun surgery (Fowler), feature envy, divergent change, Liskov substitution violation
Review Process: conventional comments (Slaughter), ship/no-ship framework, diff review vs design review, LGTM criteria
分组排列的精确术语，命名框架标注了发起人，会路由到代码评审专业知识而非通用建议，所有术语都通过了15年从业者测试。

---

Explicit Note on Packaging

打包说明

This skill uses Anthropic's native 
.skill
packaging mechanism (validate, zip, present_files) for delivery. It does NOT invoke the built-in skill creator. It replaces the built-in approach with research-backed principles from the Forge synthesis:
  • Vocabulary routing (embedding space activation via precise terminology)
  • U-shaped attention optimization (front-load vocabulary, back-load retrieval anchors)
  • Negative constraint steering (anti-patterns push past the distribution center)
  • Progressive disclosure (three-level context loading)
  • Few-shot superiority (examples beat verbose instructions)
See 
./references/skill-principles.md
for the condensed research and 
./references/skill-template.md
for an annotated gold-standard example.
本Skill使用Anthropic原生
.skill
打包机制（验证、压缩、present_files）交付，不会调用内置Skill创建器。它用来自Forge synthesis的研究支持原则替代了内置方案：
  • 词汇路由（通过精确术语在嵌入空间激活）
  • U型注意力优化（前置词汇、后置检索锚点）
  • 负向约束引导（反模式推动模型避开分布中心输出）
  • 渐进式披露（三级上下文加载）
  • 少样本优势（示例优于冗长指令）
查看
./references/skill-principles.md
获取精简研究内容，查看
./references/skill-template.md
获取带注释的黄金标准示例。

Questions This Skill Answers

本Skill回答的问题

  • "Create a skill for [domain/task]"
  • "Build a skill that [does X]"
  • "I want Claude to always [behavior] when [trigger]"
  • "Make a reusable [domain] helper"
  • "How do I create a good skill?"
  • "Turn this workflow into a skill"
  • "I need a skill for [task]"
  • "Build me a custom [domain] assistant"
  • "Create a prompt template for [task]"
  • "Package this as a skill"
  • "What makes a skill effective?"
  • "My skill isn't triggering, how do I fix it?"
  • "How should I structure a SKILL.md?"
  • "Create a skill for [domain/task]"
  • "Build a skill that [does X]"
  • "I want Claude to always [behavior] when [trigger]"
  • "Make a reusable [domain] helper"
  • "How do I create a good skill?"
  • "Turn this workflow into a skill"
  • "I need a skill for [task]"
  • "Build me a custom [domain] assistant"
  • "Create a prompt template for [task]"
  • "Package this as a skill"
  • "What makes a skill effective?"
  • "My skill isn't triggering, how do I fix it?"
  • "How should I structure a SKILL.md?"