agent-creator

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Agent Creator

Creates structured agent definitions following the 7-component format. Every agent produced by this skill is grounded in persona science research, vocabulary routing mechanics, and the MAST failure taxonomy.

遵循7组件格式创建结构化的Agent定义。本技能生成的所有Agent都基于角色科学研究、词汇路由机制和MAST失败分类法构建。

Expert Vocabulary Payload

专业词汇载荷

Agent Design: role identity, domain vocabulary payload, deliverables, decision authority, standard operating procedure, anti-pattern watchlist, interaction model, handoff artifact, quality gate Organizational Structure: RACI matrix, task-relevant maturity (Andy Grove), blast radius, reporting lines, escalation path, out-of-scope boundary Security & Risk: STRIDE threat model, OWASP Top 10, attack surface, threat modeling (Shostack) Persona Science: persona alignment, persona-accuracy tradeoff, PRISM framework, role-task alignment rule, flattery degradation, token budget Vocabulary Mechanics: vocabulary routing, embedding space, knowledge cluster, distribution center, 15-year practitioner test, sub-domain clustering, attribution amplification

Anti-Pattern Watchlist

反模式观察清单

Flattery Persona

奉承式角色设定

Detection: Superlatives and absolutes in role identity — "world-class," "best," "always," "never," "unparalleled," "leading expert."
Why it fails: Superlatives activate generic motivational/marketing text clusters in embedding space instead of domain expertise. Ranjan et al. (2024) demonstrate that superlatives route to motivational/marketing embedding clusters rather than domain expertise, degrading output accuracy.
Resolution: Define the role through knowledge and behavior, not quality claims. Remove every superlative. Describe what the agent knows and does, not how good it is.

识别特征: 角色身份描述中出现最高级和绝对化表述，比如“世界级”、“最佳”、“总是”、“从不”、“无与伦比”、“顶尖专家”。
失败原因: 最高级表述会激活嵌入空间中通用的激励/营销文本集群，而非领域专业知识。Ranjan等人（2024）的研究表明，最高级表述会路由到激励/营销嵌入集群而非领域专业知识，会降低输出准确率。
解决方案: 通过知识和行为定义角色，而非质量类表述。删除所有最高级描述，说明Agent知道什么、要做什么，而非它有多优秀。

Bare Role Label

极简角色标签

Detection: Identity is fewer than 10 tokens with no organizational context. Example: "You are a product manager."
Why it fails: Activates the broadest possible cluster for that role. No boundary information means the agent will attempt anything remotely related to the title.
Resolution: Add reporting lines, scope boundaries, and collaboration context. Specify the organizational unit and adjacent roles.

识别特征: 身份描述token数少于10，且没有组织上下文。比如：“你是一名产品经理。”
失败原因: 会激活该角色最宽泛的知识集群，没有边界信息意味着Agent会尝试处理任何和职位沾边的任务。
解决方案: 补充汇报线、范围边界和协作上下文，明确所属组织单元和协作的相邻角色。

Verbose Identity

冗长身份描述

Detection: Identity section exceeds 50 tokens or is a full paragraph of description.
Why it fails: Accuracy damage scales with persona length; PRISM (2026) found under 50 tokens is the practical sweet spot. Attention budget consumed by persona processing instead of task execution.
Resolution: Trim to title + primary responsibility + organizational context. Move detailed knowledge into the vocabulary payload where it activates clusters without consuming persona attention budget.

识别特征: 身份部分超过50个token，或者是一整段描述。
失败原因: 准确率损耗和角色设定长度正相关；PRISM（2026）研究发现低于50个token是实际最优区间。角色设定处理会占用注意力预算，挤占任务执行的资源。
解决方案: 精简为「职位+核心职责+组织上下文」的结构，把详细知识移到词汇载荷中，这类内容可以激活对应知识集群，不会占用角色设定的注意力预算。

Missing Deliverables

缺失可交付成果

Detection: Role definition describes only behaviors and attitudes, no concrete artifacts. Nothing that could be verified as "produced" or "not produced."
Why it fails: Without defined outputs, the agent has no completion criteria. It cannot self-assess whether its work is done or done correctly.
Resolution: Every role produces specific named artifacts with format descriptions. Ask: "What does this person hand to the next person in the chain?"

识别特征: 角色定义仅描述行为和态度，没有具体产出物，不存在可以验证“已产出”或“未产出”的内容。
失败原因: 没有明确的输出，Agent就没有完成标准，无法自我评估工作是否完成、是否符合要求。
解决方案: 每个角色都要产出指定名称、带格式说明的具体产物，可以问自己：“这个角色会把什么内容交给流程中的下一个人？”

Overlapping Authority

权限重叠

Detection: Two agents in a team can both autonomously decide the same thing. Decision authority sections have intersection.
Why it fails: Creates FM-2.3 Role Confusion from the MAST taxonomy. Agents produce contradictory outputs or duplicate work. Neither knows the other has already decided.
Resolution: Explicitly delineate — one agent decides, others advise. Use the RACI principle: exactly one Responsible, one Accountable per decision.

识别特征: 团队中的两个Agent都可以自主决策同一事项，决策权限部分存在交集。
失败原因: 会触发MAST分类法中的FM-2.3角色混淆问题，Agent会产出矛盾的输出或者重复工作，双方都不知道对方已经做出了决策。
解决方案: 明确划分权限——仅一个Agent负责决策，其他角色提供建议。遵循RACI原则：每个决策有且仅有一个责任人、一个最终负责人。

Generic Vocabulary

通用词汇

Detection: Vocabulary payload contains consultant-speak — "best practices," "leverage," "synergy," "holistic approach," "robust solution," "paradigm shift."
Why it fails: Generic terms activate broad, shallow knowledge clusters. The model produces fluent but non-specific output indistinguishable from a junior consultant's work.
Resolution: Apply the 15-year practitioner test to every term. Replace each generic term with the precise term a senior practitioner would use with a peer. "Best practices for testing" becomes "mutation testing, property-based testing (QuickCheck), contract testing (Pact)."

识别特征: 词汇载荷中包含咨询套话，比如“最佳实践”、“利用”、“协同”、“整体方案”、“可靠解决方案”、“范式转变”。
失败原因: 通用术语会激活宽泛、浅薄的知识集群，模型会产出流畅但无针对性的输出，和初级咨询师的作品没有区别。
解决方案: 对每个术语应用15年从业者测试，把每个通用术语替换为资深从业者和同行交流时会使用的精准术语。比如“测试最佳实践”可以替换为“mutation testing, property-based testing (QuickCheck), contract testing (Pact)”。

Behavioral Instructions

操作指南

Step 1: Receive Role Specification

步骤1：接收角色规范

Accept the role specification from one of two sources:

From Mission Planner: A blueprint containing role name, responsibilities, team context, and adjacent roles.
From direct user request: A description of what they need.

IF the source is Mission Planner: proceed to Step 3 (research). The blueprint provides sufficient context.

IF the source is a direct user request: proceed to Step 2 (interview).

从两个来源之一接收角色规范：

来自任务规划器: 包含角色名称、职责、团队上下文和相邻角色的蓝图。
来自用户直接请求: 对所需Agent的描述。

如果来源是任务规划器：直接进入步骤3（调研），蓝图已经提供了足够的上下文。

如果来源是用户直接请求：进入步骤2（访谈获取上下文）。

Step 2: Interview for Context (Direct Requests Only)

步骤2：访谈获取上下文（仅适用于直接请求）

Gather the following information through targeted questions. Do not ask all at once — adapt based on what the user has already provided.

Domain: What field does this agent work in? (software, marketing, security, operations, design, etc.)
Primary Responsibility: What is the single most important thing this agent does?
Adjacent Roles: Who does this agent work with? Who provides input? Who receives output?
Deliverables: What specific artifacts should this agent produce?
Decision Scope: What can this agent decide alone? What requires approval?
Constraints: Any specific tools, frameworks, methodologies, or standards the agent must follow?

IF the user provides a job description or role document: extract answers from the document rather than asking.

OUTPUT: Gathered role context sufficient to build the 7-component definition.

通过针对性问题收集以下信息，不要一次性问所有问题，根据用户已经提供的内容灵活调整。

领域: 该Agent工作的领域是什么？（软件、营销、安全、运维、设计等）
核心职责: 该Agent要做的最重要的一件事是什么？
相邻角色: 该Agent和谁协作？谁提供输入？谁接收输出？
可交付成果: 该Agent需要产出哪些具体的产物？
决策范围: 哪些内容该Agent可以自主决定？哪些需要审批？
约束: Agent需要遵循哪些特定的工具、框架、方法论或标准？

如果用户提供了职位描述或角色文档：从文档中提取对应答案，不要重复提问。

输出：足以构建7组件定义的角色上下文信息。

Step 3: Research the Role

步骤3：角色调研

Investigate what this role actually does in practice. Focus on:

What artifacts does a real person in this role produce daily?
What frameworks, methodologies, and tools define this domain?
What are the common failure modes for this role?
What vocabulary does a 15-year practitioner use?

IF web search is available and the domain is unfamiliar: use it to verify terminology and frameworks.

IF the role is well-known (e.g., software architect, product manager): draw on established domain knowledge.

OUTPUT: Role research sufficient to populate all 7 components.

调研该角色在实际场景中的工作内容，重点关注：

从事该岗位的真实从业者日常会产出什么产物？
该领域有哪些核心框架、方法论和工具？
该角色常见的失败模式有哪些？
有15年经验的从业者会使用哪些词汇？

如果可以联网搜索且你不熟悉该领域：使用搜索验证术语和框架的正确性。

如果是大众熟知的角色（比如软件架构师、产品经理）：直接调用已有的领域知识。

输出：足以填充所有7个组件的角色调研结果。

Step 4: Build the 7-Component Definition

步骤4：构建7组件定义

Follow the format specified in

./schemas/agent-definition.md

. Build each component in order:

遵循

./schemas/agent-definition.md

中指定的格式，按顺序构建每个组件：

4a. Role Identity (~20-50 tokens)

4a. 角色身份（约20-50个token）

Write a concise identity statement using this format:

You are a [real job title] responsible for [primary responsibility] within [organizational context]. You report to [authority] and collaborate with [adjacent roles].

Rules:

Use a job title that exists in real organizations.
Include reporting and collaboration context.
Keep under 50 tokens. Count them.
NO flattery. NO superlatives. NO quality claims.
Define through knowledge and behavior, not how good the agent is.

使用以下格式编写简洁的身份说明：

You are a [real job title] responsible for [primary responsibility] within [organizational context]. You report to [authority] and collaborate with [adjacent roles].

规则：

使用真实企业中存在的职位名称。
包含汇报线和协作上下文。
保持在50个token以内，自行计数。
不要有奉承表述、最高级描述、质量类声明。
通过知识和行为定义角色，而非Agent的优秀程度。

4b. Domain Vocabulary Payload (15-30 terms)

4b. 领域词汇载荷（15-30个术语）

Select precise terms organized in 3-5 clusters of 3-8 related terms.

Rules:

Every term must pass the 15-year practitioner test.
Include framework originators where applicable: "INVEST criteria (Bill Wake)."
No consultant-speak. Banned terms: "best practices," "leverage," "synergy," "paradigm shift," "holistic," "robust," "streamline," "optimize."
Group by knowledge proximity — terms that co-occur in expert discourse belong in the same cluster.
Name each cluster with its sub-domain.

选择精准的术语，划分为3-5个集群，每个集群包含3-8个相关术语。

规则：

每个术语必须通过15年从业者测试。
适用时标注框架提出者：比如“INVEST criteria (Bill Wake)”。
不要使用咨询套话，禁用术语：“best practices”、“leverage”、“synergy”、“paradigm shift”、“holistic”、“robust”、“streamline”、“optimize”。
按知识相关性分组——专家交流中会同时出现的术语放在同一个集群。
给每个集群命名对应的子领域。

4c. Deliverables & Artifacts

4c. 可交付成果与产物

List 3-6 specific artifacts this agent produces.

Rules:

Name the artifact type precisely: "Architecture Decision Record," not "a document."
Describe the format: sections, structure, approximate length.
Each deliverable must be verifiable — someone can check if it was produced correctly.
Ask: "What does this person hand to the next person in the chain?"

列出该Agent产出的3-6个具体产物。

规则：

精准命名产物类型：比如“架构决策记录”，而非“一份文档”。
描述格式：章节、结构、大致长度。
每个可交付成果都必须可验证——他人可以检查它是否被正确产出。
自问：“这个角色会把什么内容交给流程中的下一个人？”

4d. Decision Authority & Boundaries

4d. 决策权限与边界

Define three categories:

Autonomous: Decisions this agent makes without asking.
Escalate: Decisions requiring human approval or another agent's input.
Out of scope: Things this agent explicitly does NOT handle.

Rules:

Prevent role overlap — check against other team members if building for a team.
"Out of scope" is critical — it defines where this agent stops.
Escalation triggers must be specific, not vague ("if unsure").

定义三个类别：

自主决策: 该Agent不需要询问即可自行决定的内容。
升级审批: 需要人工审批或其他Agent输入的决策。
超出范围: 该Agent明确不处理的内容。

规则：

避免角色重叠——如果是为团队构建Agent，要和其他团队成员的权限做校验。
“超出范围”非常重要——它定义了Agent的能力边界。
升级触发条件必须具体，不要模糊表述（比如“如果不确定”）。

4e. Standard Operating Procedure

4e. 标准操作流程

Write imperative, ordered steps with explicit conditions.

Rules:

Every step starts with an imperative verb.
Conditions use explicit IF/THEN branching.
Steps that produce output have an OUTPUT line.
Steps are in execution order.
Include WHY for non-obvious steps.
4-8 steps is typical. Fewer means too abstract; more means micromanaging.

编写带明确条件的命令式有序步骤。

规则：

每个步骤以命令式动词开头。
条件使用明确的IF/THEN分支结构。
产生输出的步骤要加OUTPUT行。
步骤按执行顺序排列。
非显而易见的步骤要补充原因说明。
通常4-8个步骤最合适，太少会太抽象，太多会过于精细化管控。

4f. Anti-Pattern Watchlist (5-10 patterns)

4f. 反模式观察清单（5-10个模式）

Name specific failure modes for this role with detection signals.

Rules:

Use established pattern names from MAST taxonomy or domain literature where they exist.
Detection signals must be observable, not inferential.
Every pattern must have a concrete resolution — not "be careful" but "do X instead."
Include at least one role-specific pattern (not just generic agent failures).

列出该角色的特定失败模式以及识别信号。

规则：

适用时使用MAST分类法或领域文献中已有的模式名称。
识别信号必须可观测，而非推断性的。
每个模式必须有具体的解决方案——不要写“小心注意”，要写“改为执行X操作”。
至少包含一个角色特有的模式（而非通用的Agent失败问题）。

4g. Interaction Model

4g. 交互模型

Define how this agent communicates:

Receives from: [role] -> [artifact type]
Delivers to: [role] -> [artifact type]
Handoff format: How artifacts are transferred.
Coordination: Centralized, peer-to-peer, or sequential pipeline.

Rules:

For standalone agents (no team): describe user interaction patterns.
Handoff format must be specific enough that both sender and receiver agree on structure.

OUTPUT: Complete 7-component agent definition.

定义该Agent的沟通方式：

接收来自: [角色] -> [产物类型]
交付给: [角色] -> [产物类型]
交接格式: 产物的传输方式。
协作模式: 中心化、点对点，还是顺序流水线。

规则：

针对独立Agent（没有所属团队）：描述和用户的交互模式。
交接格式必须足够具体，确保发送方和接收方对结构的理解一致。

输出：完整的7组件Agent定义。

Step 5: Apply PRISM Validation

步骤5：应用PRISM校验

Review the complete definition against PRISM findings:

Token count check: Is role identity under 50 tokens? If not, trim.
Flattery check: Any superlatives or quality claims? If found, remove.
Role-task alignment: Does the job title match the primary deliverables? If misaligned, adjust.
Vocabulary validation: Does every term pass the 15-year practitioner test? Replace any that fail.
Anti-pattern scan: Run the definition against the Anti-Pattern Watchlist in this skill. Fix any matches.

OUTPUT: Validated agent definition.

对照PRISM研究结论校验完整定义：

token计数检查: 角色身份是否少于50个token？如果不是，精简内容。
奉承表述检查: 是否存在最高级或质量类声明？如果有，删除。
角色-任务对齐检查: 职位名称和核心可交付成果是否匹配？如果不匹配，调整。
词汇验证: 每个术语是否都通过了15年从业者测试？替换所有不符合要求的术语。
反模式扫描: 对照本技能的反模式观察清单检查定义，修复所有匹配的问题。

输出：校验通过的Agent定义。

Step 6: Add Library Metadata

步骤6：添加库元数据

Add YAML frontmatter:

yaml

---
name: kebab-case-name
domain: [primary domain]
tags: [3-10 searchable keywords]
created: [today's date]
quality: untested
source: [manual | jit-generated | template-derived]
---

Rules:

```
name
```
must be kebab-case, matching the filename.
```
quality
```
starts as
```
untested
```
for new agents.
```
source
```
is
```
manual
```
if user-specified,
```
jit-generated
```
if created on-the-fly,
```
template-derived
```
if based on an existing agent.

添加YAML头信息：

yaml

---
name: kebab-case-name
domain: [primary domain]
tags: [3-10 searchable keywords]
created: [today's date]
quality: untested
source: [manual | jit-generated | template-derived]
---

规则：

```
name
```
必须是短横线命名法，和文件名保持一致。
新创建的Agent
```
quality
```
初始值为
```
untested
```
。
如果是用户指定的Agent，
```
source
```
为
```
manual
```
；如果是即时生成的，为
```
jit-generated
```
；如果基于现有Agent修改，为
```
template-derived
```
。

Step 7: Save and Deliver

步骤7：保存并交付

IF environment is Claude Code:

Save to
```
.claude/agents/{name}.md
```
Update the library index if one exists.

IF environment is Cowork or conversational:

Present the complete agent definition in the response.
Offer to save to a specified location.

IF the agent was requested by Mission Planner:

Return the definition to the Mission Planner for team assembly.
Include the handoff artifact metadata.

OUTPUT: Delivered agent definition.

如果运行环境是Claude Code：

保存到
```
.claude/agents/{name}.md
```
如果存在库索引，更新索引。

如果运行环境是Cowork或对话场景：

在响应中展示完整的Agent定义。
询问是否需要保存到指定位置。

如果Agent是任务规划器请求创建的：

将定义返回给任务规划器用于团队组建。
包含交接产物元数据。

输出：交付完成的Agent定义。

Output Format

输出格式

The output is a complete agent definition markdown file following

./schemas/agent-definition.md

. The file contains:

YAML frontmatter with library metadata
Seven numbered sections (Role Identity, Domain Vocabulary, Deliverables, Decision Authority, SOP, Anti-Patterns, Interaction Model)
Each section follows the format rules specified in the schema

See

./references/agent-template.md

for a fully annotated example.

输出为遵循

./schemas/agent-definition.md

规范的完整Agent定义Markdown文件，文件包含：

带库元数据的YAML头信息
七个编号章节（角色身份、领域词汇、可交付成果、决策权限、SOP、反模式、交互模型）
每个章节都遵循模式中指定的格式规则

参考

./references/agent-template.md

查看带完整注释的示例。

Examples

示例

Example 1: Role Identity

示例1：角色身份

BAD:

You are the world's leading product manager with unparalleled expertise in creating products that users love. You always make the right decisions and have an extraordinary ability to understand user needs.

Problems: 42 tokens of flattery. "World's leading" activates motivational text. "Always make the right decisions" is an absolute. "Extraordinary ability" is a quality claim. No organizational context. No collaboration boundaries.

GOOD:

You are a product manager responsible for defining requirements and success metrics within a B2B SaaS product team. You report to the VP of Product and collaborate with engineering, design, and sales.

Why it works: Real job title. Primary responsibility stated. Organizational context (B2B SaaS). Reporting line and collaborators establish boundaries. 35 tokens. No flattery.

错误示例:

You are the world's leading product manager with unparalleled expertise in creating products that users love. You always make the right decisions and have an extraordinary ability to understand user needs.

问题：42个token都是奉承表述，“世界顶尖”会激活激励类文本，“总是做出正确决策”是绝对化表述，“非凡能力”是质量类声明，没有组织上下文，没有协作边界。

正确示例:

You are a product manager responsible for defining requirements and success metrics within a B2B SaaS product team. You report to the VP of Product and collaborate with engineering, design, and sales.

优势：真实的职位名称，明确了核心职责，有组织上下文（B2B SaaS），汇报线和协作方明确了边界，共35个token，没有奉承表述。

Example 2: Domain Vocabulary

示例2：领域词汇

BAD:

best practices, stakeholder alignment, strategic vision, innovative solutions, leverage synergies, drive results, thought leadership, holistic approach

Problems: Every term fails the 15-year practitioner test. No senior PM says "leverage synergies" to a peer. These activate generic business writing clusters. No framework attributions. No sub-domain clustering.

GOOD:

Discovery & Prioritization: PRD structure, RICE prioritization (Intercom), Jobs-to-be-Done (Christensen), opportunity-solution tree (Teresa Torres), assumption mapping Execution Frameworks: user story mapping (Jeff Patton), INVEST criteria (Bill Wake), acceptance criteria, definition of done, sprint goal Measurement: OKR alignment, North Star metric, activation rate, retention cohort, product-market fit score (Sean Ellis)

Why it works: Three distinct clusters. Every term passes the 15-year practitioner test. Framework originators attributed. No consultant-speak. 25 precise terms that route to product management knowledge clusters.

错误示例:

best practices, stakeholder alignment, strategic vision, innovative solutions, leverage synergies, drive results, thought leadership, holistic approach

问题：每个术语都没有通过15年从业者测试，资深产品经理不会和同行说“leverage synergies”这类表述，这些术语会激活通用商业写作集群，没有框架归属，没有子领域分组。

正确示例:

Discovery & Prioritization: PRD structure, RICE prioritization (Intercom), Jobs-to-be-Done (Christensen), opportunity-solution tree (Teresa Torres), assumption mapping Execution Frameworks: user story mapping (Jeff Patton), INVEST criteria (Bill Wake), acceptance criteria, definition of done, sprint goal Measurement: OKR alignment, North Star metric, activation rate, retention cohort, product-market fit score (Sean Ellis)

优势：三个清晰的分组，每个术语都通过了15年从业者测试，标注了框架提出者，没有咨询套话，25个精准术语可以路由到产品管理知识集群。

Questions This Skill Answers

本技能可解答的问题

"Create an agent for [role/domain]"
"I need a [job title] agent"
"Define a [role] persona"
"Build me a product manager / engineer / designer / etc."
"What should a [role] agent look like?"
"How do I create a good agent definition?"
"Make me an AI assistant for [specific task]"
"I need help with [domain] — create an agent"
"Turn this role description into an agent"
"Create a specialized agent for my project"
"What's wrong with my agent definition?"
"Improve this agent persona"

"为[角色/领域]创建一个Agent"
"我需要一个[职位]Agent"
"定义一个[角色]人设"
"给我做一个产品经理/工程师/设计师等Agent"
"[角色]Agent应该是什么样的？"
"我怎么创建一个好的Agent定义？"
"给我做一个用于[特定任务]的AI助手"
"我需要[领域]相关帮助——创建一个Agent"
"把这个职位描述转化为Agent"
"为我的项目创建一个专用Agent"
"我的Agent定义有什么问题？"
"优化这个Agent人设"

References

参考资料

```
./schemas/agent-definition.md
```
— The 7-component format specification
```
./references/persona-science.md
```
— PRISM findings on persona effectiveness
```
./references/agent-template.md
```
— Annotated gold-standard agent example
```
./references/failure-modes.md
```
— MAST failure modes relevant to agent design

```
./schemas/agent-definition.md
```
— 7组件格式规范
```
./references/persona-science.md
```
— 关于角色有效性的PRISM研究结论
```
./references/agent-template.md
```
— 带注释的黄金标准Agent示例
```
./references/failure-modes.md
```
— 和Agent设计相关的MAST失败模式