blog-persona

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Blog Persona - Writing Voice Management

博客写作角色 - 语气管理

Create, store, and enforce writing personas based on the NNGroup 4-dimension tone framework and CMI Brand Voice Chart. Personas ensure consistent voice across all blog content produced by blog-write and blog-rewrite.

基于NNGroup四维语气框架和CMI品牌语气图表创建、存储并执行写作角色。写作角色可确保blog-write和blog-rewrite生成的所有博客内容语气一致。

Commands

命令

Command	Purpose
`/blog persona create`	Interactive interview to build a new persona
`/blog persona list`	Show all saved personas
`/blog persona use <name>`	Set active persona for current session
`/blog persona show <name>`	Display full persona profile

命令	用途
`/blog persona create`	交互式访谈以创建新角色
`/blog persona list`	显示所有已保存的角色
`/blog persona use <name>`	设置当前会话的活跃角色
`/blog persona show <name>`	显示完整的角色档案

Create Workflow

创建流程

Run the 6-step interactive interview. Ask each step, wait for response, then proceed.

运行6步交互式访谈。依次询问每一步，等待用户回复后再继续。

Step 1: Brand Basics

步骤1：品牌基础

Ask the user for:

Brand name - company or personal brand
Industry - primary sector (e.g., SaaS, health, finance, education)
Target audience - who reads the blog (role, experience level, goals)
One-sentence brand mission - what the brand helps people do

询问用户以下信息：

品牌名称 - 企业或个人品牌
行业 - 核心领域（如SaaS、健康、金融、教育）
目标受众 - 博客的读者群体（角色、经验水平、目标）
一句话品牌使命 - 品牌能为用户提供什么帮助

Step 2: Tone Dimensions (NNGroup Framework)

步骤2：语气维度（NNGroup框架）

Present each dimension as a 0.0 to 1.0 slider. Explain both ends with examples.

Dimension	0.0 End	1.0 End	Example at 0.0	Example at 1.0
funny_serious	Funny	Serious	"Let's be real, nobody reads Terms of Service"	"Understanding legal agreements protects your business"
formal_casual	Formal	Casual	"We are pleased to announce"	"Guess what - we shipped it!"
respectful_irreverent	Respectful	Irreverent	"We appreciate your patience"	"Yeah, that old way was broken"
enthusiastic_matter_of_fact	Enthusiastic	Matter-of-fact	"This changes everything!"	"Here are the results."

Defaults if user is unsure:

[0.6, 0.5, 0.3, 0.5]

(slightly serious, balanced formality, respectful, balanced enthusiasm).

将每个维度展示为0.0到1.0的滑块，并用示例解释两端的含义。

维度	0.0端	1.0端	0.0端示例	1.0端示例
funny_serious	诙谐	严肃	"说实话，没人会读服务条款"	"理解法律协议能保护您的业务"
formal_casual	正式	随意	"我们很高兴宣布"	"猜猜看——我们上线了！"
respectful_irreverent	恭敬	不敬	"感谢您的耐心等待"	"没错，旧方法已经行不通了"
enthusiastic_matter_of_fact	热情	实事求是	"这彻底改变了一切！"	"以下是结果。"

若用户不确定，默认值为：

[0.6, 0.5, 0.3, 0.5]

（略带严肃、正式与随意平衡、恭敬、热情与实事求是平衡）。

Step 3: Writing Rules

步骤3：写作规则

Ask the user to pick a vocabulary tier first, then auto-suggest the matching readability band (user can override).

Setting	What to Ask	Default
Vocabulary tier	Consumer, Professional, or Technical	Professional
Readability band	Auto-filled from tier (see table below)	Grade 8-10
Sentence length mean	Average words per sentence	18
Sentence length std	Variation in sentence length	6
Contraction frequency	0.0 (never) to 1.0 (always)	0.6
Max passive voice	Percentage cap on passive constructions	10%

先询问用户选择词汇层级，然后自动推荐匹配的可读性区间（用户可覆盖）。

设置	询问内容	默认值
词汇层级	大众级、专业级或技术级	专业级
可读性区间	根据层级自动填充（见下表）	8-10年级水平
平均句子长度	平均每句单词数	18
句子长度标准差	句子长度的变化幅度	6
缩写使用频率	0.0（从不）至1.0（总是）	0.6
被动语态上限	被动句式的占比上限	10%

Step 4: Do's and Don'ts (CMI Brand Voice Chart)

步骤4：注意事项（CMI品牌语气图表）

Ask for 3-5 items in each list. Provide starter examples based on the tone dimensions.

Example Do's: "Use data to back claims", "Address the reader as you", "Open with a question or stat"

Example Don'ts: "Don't use jargon without defining it", "Don't start sentences with There is/There are", "Don't use cliches like game-changer"

询问用户在每个列表中提供3-5项内容。根据语气维度提供示例参考。

示例注意事项（Do）： "用数据支撑观点"、"用‘你’称呼读者"、"以问题或数据开篇"

示例注意事项（Don't）： "不要使用未定义的行话"、"不要以‘There is/There are’开头"、"不要使用‘game-changer’这类陈词滥调"

Step 5: Summary Label Preference

步骤5：摘要标签偏好

The label used for summary/takeaway boxes in blog posts. Ask user to pick one:

Key Takeaways (default)
The Bottom Line
What You'll Learn
TL;DR
Quick Summary
In a Nutshell
Custom label

博客文章中摘要/要点框使用的标签。让用户选择以下选项之一：

Key Takeaways（默认）
The Bottom Line
What You'll Learn
TL;DR
Quick Summary
In a Nutshell
自定义标签

Step 6: Voice Samples (Optional)

步骤6：语气示例（可选）

Ask if the user has 1-3 URLs of existing content that exemplifies the desired voice. Store URLs in the persona for future reference. If provided, read each URL and extract:

Average sentence length
Contraction frequency
Tone dimension estimates
Vocabulary level

Compare extracted values with the persona settings and flag any mismatches.

询问用户是否有1-3个体现理想语气的现有内容URL。将URL存储在角色档案中供后续参考。若提供URL，读取每个URL并提取：

平均句子长度
缩写使用频率
语气维度估算值
词汇水平

将提取的值与角色设置进行比较，标记任何不匹配之处。

Save

保存

Write the completed persona as JSON to:

skills/blog/references/personas/<name>.json

Use kebab-case for the filename (e.g.,

acme-saas.json

将完成的角色以JSON格式写入：

skills/blog/references/personas/<name>.json

文件名使用短横线分隔的小写格式（如

acme-saas.json

）。

Persona Profile Schema

角色档案架构

json

{
  "name": "acme-saas",
  "description": "Professional SaaS voice for B2B marketing content",
  "brand": "Acme Corp",
  "industry": "SaaS",
  "audience": "Marketing managers at mid-market companies",
  "mission": "Help marketing teams automate reporting",
  "tone_dimensions": {
    "funny_serious": 0.7,
    "formal_casual": 0.4,
    "respectful_irreverent": 0.2,
    "enthusiastic_matter_of_fact": 0.5
  },
  "readability": {
    "flesch_grade_min": 8,
    "flesch_grade_max": 10,
    "flesch_ease_min": 50,
    "flesch_ease_max": 60
  },
  "style": {
    "sentence_length_mean": 18,
    "sentence_length_std": 6,
    "contraction_frequency": 0.6,
    "passive_voice_max_pct": 10,
    "vocabulary_tier": "professional",
    "summary_label": "Key Takeaways"
  },
  "voice_samples": [],
  "do": [
    "Use data to back every major claim",
    "Address the reader directly as you",
    "Lead sections with actionable insight"
  ],
  "dont": [
    "Don't use buzzwords without context",
    "Don't write sentences longer than 30 words",
    "Don't open with We at Acme"
  ]
}

json

{
  "name": "acme-saas",
  "description": "Professional SaaS voice for B2B marketing content",
  "brand": "Acme Corp",
  "industry": "SaaS",
  "audience": "Marketing managers at mid-market companies",
  "mission": "Help marketing teams automate reporting",
  "tone_dimensions": {
    "funny_serious": 0.7,
    "formal_casual": 0.4,
    "respectful_irreverent": 0.2,
    "enthusiastic_matter_of_fact": 0.5
  },
  "readability": {
    "flesch_grade_min": 8,
    "flesch_grade_max": 10,
    "flesch_ease_min": 50,
    "flesch_ease_max": 60
  },
  "style": {
    "sentence_length_mean": 18,
    "sentence_length_std": 6,
    "contraction_frequency": 0.6,
    "passive_voice_max_pct": 10,
    "vocabulary_tier": "professional",
    "summary_label": "Key Takeaways"
  },
  "voice_samples": [],
  "do": [
    "Use data to back every major claim",
    "Address the reader directly as you",
    "Lead sections with actionable insight"
  ],
  "dont": [
    "Don't use buzzwords without context",
    "Don't write sentences longer than 30 words",
    "Don't open with We at Acme"
  ]
}

Readability Bands by Vocabulary Tier

各词汇层级对应的可读性区间

Tier	Flesch Grade	Flesch Ease	Typical Use
Consumer	6-8	60-80	Health, lifestyle, personal finance
Professional	8-10	50-60	B2B, marketing, management
Technical	10-12	30-50	Engineering, medical, legal

When the user picks a tier, auto-fill the readability fields. Let them override if they want a non-standard combination (e.g., technical vocabulary at consumer readability for explainer content).

层级	Flesch年级水平	Flesch易读度	典型用途
大众级	6-8	60-80	健康、生活方式、个人理财
专业级	8-10	50-60	B2B、营销、管理
技术级	10-12	30-50	工程、医疗、法律

当用户选择层级后，自动填充可读性字段。若用户需要非标准组合（如面向大众可读性的技术词汇解释内容），允许其覆盖默认值。

Integration with blog-write and blog-rewrite

与blog-write和blog-rewrite的集成

When a persona is active (via

/blog persona use <name>

), the writer agent loads the persona JSON and enforces these constraints during generation:

Pre-generation - Load persona, inject tone dimensions and style rules into the system prompt for the blog-writer agent.
During generation - Writer follows do/dont rules, targets sentence length mean/std, uses contractions at specified frequency.
Post-generation validation - Check the output against persona constraints:
- Sentence length distribution within 1 std of target mean
- Readability score within the specified grade band
- Passive voice percentage under the max
- No violations of "dont" rules found via pattern matching

If validation fails, flag the specific violations and suggest edits.

当角色处于活跃状态（通过

/blog persona use <name>

命令）时，写作Agent会加载角色JSON，并在生成内容时执行以下约束：

生成前 - 加载角色，将语气维度和风格规则注入blog-writer Agent的系统提示词中。
生成过程中 - 写作遵循Do/Don't规则，以目标平均句子长度/标准差为基准，按指定频率使用缩写。
生成后验证 - 检查输出是否符合角色约束：
- 句子长度分布在目标平均值的1个标准差范围内
- 可读性分数在指定年级区间内
- 被动语态占比低于上限
- 通过模式匹配未发现违反Don't规则的内容

若验证失败，标记具体违规内容并建议修改。

List Command

列表命令

Glob

skills/blog/references/personas/*.json

and display a table:

Persona	Industry	Audience	Vocabulary
acme-saas	SaaS	Marketing managers	Professional

If no personas exist, prompt the user to create one.

遍历

skills/blog/references/personas/*.json

文件并显示表格：

角色	行业	受众	词汇层级
acme-saas	SaaS	营销经理（中型企业）	专业级

若不存在任何角色，提示用户创建一个。

Show Command

展示命令

Read the specified persona JSON and display it as a formatted summary with all tone dimensions, style rules, and do/dont lists.

读取指定的角色JSON文件，以格式化摘要的形式显示所有语气维度、风格规则和Do/Don't列表。

Use Command

使用命令

Read the persona JSON and confirm activation. Print a summary of the key constraints that will be enforced. The persona stays active for the current conversation session. Blog-write and blog-rewrite check for the active persona before generating content.

读取角色JSON并确认激活。打印将执行的关键约束摘要。该角色在当前对话会话中保持活跃状态。blog-write和blog-rewrite在生成内容前会检查是否有活跃角色。

Error Handling

错误处理

Invalid tone values: If a user provides values outside 0.0-1.0, clamp to the nearest valid bound and warn
Unreachable voice samples: If a URL in voice_samples returns an error, skip it and note in the profile that the sample was unavailable
Empty personas directory: When running list or show with no personas saved, prompt the user to create one first
Name conflicts: If a persona name already exists during create, ask whether to overwrite or choose a different name
Malformed JSON: If a persona file is corrupted, report the error and offer to recreate it from the interview

无效语气值：若用户提供的值超出0.0-1.0范围，将其限制为最近的有效值并发出警告
无法访问的语气示例：若voice_samples中的URL返回错误，跳过该URL并在档案中注明示例不可用
角色目录为空：当运行列表或展示命令但未保存任何角色时，提示用户先创建一个
名称冲突：创建角色时若名称已存在，询问用户是否覆盖或选择其他名称
JSON格式错误：若角色文件损坏，报告错误并提供通过访谈重新创建的选项