blog-cluster

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Blog Cluster (Semantic Topic Cluster Engine)

博客集群（语义主题集群引擎）

Plans and executes entire interlinked content ecosystems from a single seed keyword. Three layers: Semantic Clustering (the brain), Cluster Architecture (the structure), and Execution Engine (the machine).

Adapted from the semantic-cluster-engine submission by Lutfiya Miller (winner, AI Marketing Hub Pro Challenge, March 2026, 95/100 Exemplary). Original repository: https://github.com/Drfiya/semantic-cluster-engine This port keeps the Plan + Execute architecture and the cluster context innovation, removes brand-specific (ScienceExperts.ai) styling and image prompts, and routes through claude-blog's existing sub-skills.

从单个种子关键词规划并执行完整的互联内容生态系统。包含三层架构：语义聚类（核心大脑）、集群架构（结构框架）和执行引擎（执行机制）。

改编自Lutfiya Miller提交的semantic-cluster-engine （2026年3月AI Marketing Hub专业挑战赛冠军，评分95/100，获“优秀”评级）。原仓库地址：https://github.com/Drfiya/semantic-cluster-engine 本次移植保留了“规划+执行”架构和集群上下文创新，移除了特定品牌（ScienceExperts.ai）的样式和图片提示，并通过claude-blog的现有子技能进行路由。

Commands

命令

Command	What it does
`/blog cluster`	Interactive. Asks whether to plan or execute.
`/blog cluster plan <seed-keyword>`	SERP-based semantic analysis. Outputs cluster plan + map.
`/blog cluster plan --from strategy [path]`	Imports existing `blog-strategy` cluster build plan and validates against SERP data.
`/blog cluster execute [path-to-plan]`	Sequential `blog-write` calls with cluster context and auto-interlinks.

命令	功能说明
`/blog cluster`	交互式命令。询问用户是要规划集群还是执行现有规划。
`/blog cluster plan <seed-keyword>`	基于SERP的语义分析。输出集群规划方案和集群地图。
`/blog cluster plan --from strategy [path]`	导入已有的 `blog-strategy` 集群构建规划，并基于SERP数据进行验证。
`/blog cluster execute [path-to-plan]`	按顺序调用 `blog-write` ，附带集群上下文并自动添加内部链接。

Key references (load on demand)

关键参考资料（按需加载）

```
references/semantic-clustering.md
```
(SERP overlap analysis, intent classification, keyword universe expansion)
```
references/cluster-architecture.md
```
(hub-and-spoke specs, schema strategy, link-density rules)
```
references/execution-workflow.md
```
(execution order, context injection, scorecard, failure handling)

```
references/semantic-clustering.md
```
（SERP重叠分析、意图分类、关键词范围扩展）
```
references/cluster-architecture.md
```
（Hub-and-Spoke规范、Schema策略、链接密度规则）
```
references/execution-workflow.md
```
（执行顺序、上下文注入、评分卡、故障处理）

Cross-references to existing claude-blog skills

与现有claude-blog技能的交叉引用

Skill	When this skill calls it
`/blog strategy`	Upstream planning. `plan --from strategy` consumes its Cluster Build Plan tables.
`/blog write`	Per-post execution. Each spoke and the pillar are produced by `blog-write` with a prepended cluster-context block.
`/blog chart`	Invoked internally by `blog-write` for inline SVG charts. No direct call from this skill.
`/blog image`	Optional hero image generation per post (graceful fallback if `nanobanana-mcp` is not configured).
`/blog seo-check`	Recommended after execution for per-post on-page validation.
`/blog cannibalization`	Recommended after execution to confirm zero keyword overlap across the cluster.
`/blog schema`	Recommended after execution to add `BreadcrumbList` , `ItemList` , and `Article` schema.

This skill never modifies files belonging to other skills. It calls them via the Task tool or as orchestrated sub-skills.

技能	调用时机
`/blog strategy`	上游规划阶段。 `plan --from strategy` 会调用其生成的集群构建规划表格。
`/blog write`	单篇内容执行阶段。每个分支内容和支柱内容都由 `blog-write` 生成，且会预先添加集群上下文块。
`/blog chart`	由 `blog-write` 内部调用以生成嵌入式SVG图表。本技能不会直接调用它。
`/blog image`	可选功能，为每篇文章生成封面图（若未配置 `nanobanana-mcp` 则自动降级处理）。
`/blog seo-check`	执行完成后推荐使用，用于验证单篇内容的页面SEO情况。
`/blog cannibalization`	执行完成后推荐使用，用于确认集群内无关键词重叠问题。
`/blog schema`	执行完成后推荐使用，用于添加 `BreadcrumbList` 、 `ItemList` 和 `Article` 类型的Schema。

本技能不会修改其他技能所属的文件，仅通过任务工具或编排子技能的方式调用它们。

Command Routing

命令路由逻辑

Parse the user's command to determine the sub-command.
If the user typed only
```
/blog cluster
```
, ask: "Would you like to plan a new cluster or execute an existing plan?"
Route:
- ```
plan <keyword>
```
  to the Plan Phase (below)
- ```
plan --from strategy [path]
```
  to the Strategy Import flow (below)
- ```
execute [path]
```
  ,
```
build
```
  , or
```
run
```
  to the Execute Phase (below)

解析用户命令以确定子命令类型。
如果用户仅输入
```
/blog cluster
```
，则询问：“您想要规划新集群还是执行现有规划？”
路由规则：
- ```
plan <keyword>
```
  进入规划阶段（下文详述）
- ```
plan --from strategy [path]
```
  进入策略导入流程（下文详述）
- ```
execute [path]
```
  、
```
build
```
  或
```
run
```
  进入执行阶段（下文详述）

Plan Phase:

/blog cluster plan <seed-keyword>

规划阶段：

/blog cluster plan <seed-keyword>

Reference:

references/semantic-clustering.md

参考资料：

references/semantic-clustering.md

Step 1. Seed keyword expansion

步骤1：种子关键词扩展

Use WebSearch to expand the seed into a keyword universe of 30 to 50 phrases:

Direct search of
```
<seed>
```
to capture related searches and "People also ask".

Long-tail expansion:

<seed> guide

<seed> tips

<seed> tools

<seed> examples

<seed> vs

best <seed>

how to <seed>

Question mining:

what is <seed>

how does <seed> work

why <seed>

<seed> for beginners

Intent variants: add commercial modifiers (best, top, review, comparison, pricing), informational modifiers (guide, tutorial, explained, examples), and transactional modifiers (buy, download, tool, software, service).
Year freshness:
```
<seed> 2026
```
.

使用WebSearch将种子关键词扩展为包含30至50个短语的关键词库：

直接搜索
```
<seed>
```
以获取相关搜索结果和“People also ask”内容。

长尾词扩展：

<seed> guide

、

<seed> tips

、

<seed> tools

、

<seed> examples

、

<seed> vs

、

best <seed>

、

how to <seed>

。

问题挖掘：

what is <seed>

、

how does <seed> work

、

why <seed>

、

<seed> for beginners

。

意图变体：添加商业类修饰词（best、top、review、comparison、pricing）、信息类修饰词（guide、tutorial、explained、examples）和交易类修饰词（buy、download、tool、software、service）。
时效性优化：
```
<seed> 2026
```
。

Step 2. Semantic clustering

步骤2：语义聚类

Group the expanded keywords using the priority rules in

references/semantic-clustering.md

SERP Overlap Analysis is the primary signal. Two keywords with 5 or more shared top-10 results target the same intent and belong in one post.
Intent Classification assigns each keyword to informational, commercial, transactional, or navigational.
Entity Mapping identifies the people, products, frameworks, and organizations Google associates with the topic.
Grouping combines keywords that share intent and topical proximity. Each group becomes one branch of the hub and spoke.

依据

references/semantic-clustering.md

中的优先级规则对扩展后的关键词进行分组：

SERP重叠分析是核心判断依据。若两个关键词的前10条搜索结果中有5条及以上重合，则说明它们目标意图一致，应归为同一篇内容。
意图分类：将每个关键词归类为信息型、商业型、交易型或导航型。
实体映射：识别谷歌与该主题关联的人物、产品、框架和组织。
分组：将意图和主题相关性一致的关键词合并。每个组将成为Hub-and-Spoke架构的一个分支。

Step 3. Cluster architecture design

步骤3：集群架构设计

Reference:

references/cluster-architecture.md

Build the hub and spoke:

Pillar (hub): targets the broadest keyword. Word count 2,500 to 4,000. Template
```
pillar-page
```
. Links down to every spoke.
Spokes: each targets a long-tail cluster. Word count 1,200 to 1,800. Template auto-selected by intent. Links up to the pillar and across to siblings.

Cluster formation rules:

2 to 5 clusters per pillar.
2 to 4 spokes per cluster.
Total: 1 pillar plus 5 to 15 spokes.
Every spoke targets a unique primary keyword (zero cannibalization).

参考资料：

references/cluster-architecture.md

构建Hub-and-Spoke架构：

支柱内容（Hub）：针对最宽泛的关键词。字数要求2500至4000字，使用
```
pillar-page
```
模板。链接到所有分支内容。
分支内容（Spokes）：每个内容针对一组长尾关键词。字数要求1200至1800字，根据意图自动选择模板。链接到支柱内容和同组内的其他分支内容。

集群形成规则：

每个支柱内容对应2至5个集群。
每个集群包含2至4个分支内容。
总内容量：1篇支柱内容 + 5至15篇分支内容。
每个分支内容对应唯一的主关键词（避免关键词 cannibalization）。

Step 4. Internal link matrix

步骤4：内部链接矩阵

For each spoke

```
S
```
to Pillar (always; anchor text uses the pillar's primary keyword).
Pillar to
```
S
```
(always; anchor text uses
```
S
```
's primary keyword).
```
S
```
to other spokes in the same cluster (2 to 3 links each, contextual anchors).
```
S
```
to spokes in adjacent clusters (0 to 1 links, only when semantically relevant).

Verify every spoke has at least 3 incoming links. Count total planned interlinks.

对于每个分支内容

：

```
S
```
链接到支柱内容（必选；锚文本使用支柱内容的主关键词）。
支柱内容链接到
```
S
```
（必选；锚文本使用
```
S
```
的主关键词）。
```
S
```
链接到同集群内的其他分支内容（每个链接2至3个，使用上下文锚文本）。
```
S
```
链接到相邻集群的分支内容（0至1个，仅当语义相关时添加）。

验证每个分支内容至少有3个入站链接，并统计计划的总内部链接数。

Step 5. Generate output files

步骤5：生成输出文件

All plan and execute artifacts go into a single subdirectory of the current working directory:

<cwd>/
└── cluster-<seed-keyword-slug>/
    ├── cluster-plan.json
    ├── cluster-map.html
    ├── pillar-<slug>.md       (Execute Phase)
    ├── <spoke-slug>.md        (Execute Phase, one per spoke)
    └── cluster-scorecard.md   (Execute Phase)

所有规划和执行产物将存入当前工作目录的一个子目录中：

<cwd>/
└── cluster-<seed-keyword-slug>/
    ├── cluster-plan.json
    ├── cluster-map.html
    ├── pillar-<slug>.md       （执行阶段生成）
    ├── <spoke-slug>.md        （执行阶段生成，每个分支对应一个文件）
    └── cluster-scorecard.md   （执行阶段生成）

cluster-plan.json

schema

cluster-plan.json

架构

json

{
  "seed_keyword": "<seed>",
  "generated_at": "YYYY-MM-DDTHH:MM:SSZ",
  "pillar": {
    "id": "P",
    "title": "Title of the pillar",
    "primary_keyword": "broadest keyword",
    "secondary_keywords": ["..."],
    "search_volume_estimate": "high|medium|low",
    "template": "pillar-page",
    "word_count_target": 3000,
    "cluster": "pillar"
  },
  "clusters": [
    {
      "name": "Cluster A: Theme",
      "intent": "informational|commercial|transactional",
      "color": "#2563eb",
      "posts": [
        {
          "id": "A1",
          "title": "Post title",
          "primary_keyword": "long-tail keyword",
          "secondary_keywords": ["..."],
          "search_volume_estimate": "high|medium|low",
          "template": "how-to-guide",
          "word_count_target": 1500,
          "links_to": ["P", "A2"],
          "links_from": ["P", "A2"]
        }
      ]
    }
  ],
  "total_posts": 9,
  "total_interlinks": 23,
  "estimated_total_words": 18000
}

Note: volume estimates are relative indicators (high, medium, low) derived from SERP signals, not absolute search volumes. For precise data, the user should consult Ahrefs, SEMrush, or DataForSEO (claude-blog provides the

seo-dataforseo

companion sibling).

json

{
  "seed_keyword": "<seed>",
  "generated_at": "YYYY-MM-DDTHH:MM:SSZ",
  "pillar": {
    "id": "P",
    "title": "Title of the pillar",
    "primary_keyword": "broadest keyword",
    "secondary_keywords": ["..."],
    "search_volume_estimate": "high|medium|low",
    "template": "pillar-page",
    "word_count_target": 3000,
    "cluster": "pillar"
  },
  "clusters": [
    {
      "name": "Cluster A: Theme",
      "intent": "informational|commercial|transactional",
      "color": "#2563eb",
      "posts": [
        {
          "id": "A1",
          "title": "Post title",
          "primary_keyword": "long-tail keyword",
          "secondary_keywords": ["..."],
          "search_volume_estimate": "high|medium|low",
          "template": "how-to-guide",
          "word_count_target": 1500,
          "links_to": ["P", "A2"],
          "links_from": ["P", "A2"]
        }
      ]
    }
  ],
  "total_posts": 9,
  "total_interlinks": 23,
  "estimated_total_words": 18000
}

注：搜索量估算为基于SERP信号的相对指标（高/中/低），而非绝对搜索量。如需精确数据，用户应参考Ahrefs、SEMrush或DataForSEO（claude-blog提供

seo-dataforseo

配套子技能）。

cluster-map.html

(XSS-safe)

cluster-map.html

（防XSS安全）

A static, self-contained HTML file with an embedded SVG visualization. Hard rules for the writer:

No inline
```
<script>
```
blocks. No
```
onclick
```
,
```
onmouseover
```
, or any
```
on*
```
event attributes anywhere in the document.
No external script
```
<src>
```
references.
Every text label drawn into the SVG (titles, keywords, cluster names) must be escaped: replace
```
&
```
with
```
&amp;
```
,
```
<
```
with
```
&lt;
```
,
```
>
```
with
```
&gt;
```
,
```
"
```
with
```
&quot;
```
, and
```
'
```
with
```
&#39;
```
before insertion.
Hover effects use CSS
```
:hover
```
only. No JavaScript.
Use
```
<title>
```
child elements inside SVG nodes for accessible tooltips (browser native, no script).

The map shows: a central pillar node, color-coded cluster groups radiating outward, spoke nodes within each cluster, and link lines connecting related nodes.

一个静态的独立HTML文件，包含嵌入式SVG可视化。编写时需遵循严格规则：

禁止使用内联
```
<script>
```
块。文档中任何位置不得使用
```
onclick
```
、
```
onmouseover
```
或任何
```
on*
```
事件属性。
禁止引用外部脚本
```
<src>
```
。
SVG中绘制的所有文本标签（标题、关键词、集群名称）必须转义：插入前将
```
&
```
替换为
```
&amp;
```
，
```
<
```
替换为
```
&lt;
```
，
```
>
```
替换为
```
&gt;
```
，
```
"
```
替换为
```
&quot;
```
，
```
'
```
替换为
```
&#39;
```
。
悬停效果仅使用CSS
```
:hover
```
实现，禁止使用JavaScript。
使用SVG节点内的
```
<title>
```
子元素实现可访问的工具提示（浏览器原生支持，无需脚本）。

该地图展示：中心支柱节点、向外辐射的颜色编码集群组、每个集群内的分支节点，以及连接相关节点的链接线条。

Step 6. Present plan to user

步骤6：向用户展示规划方案

Show a summary table of clusters and posts, total interlinks, estimated words, and the file paths. Ask for confirmation before proceeding to execution. Wait for explicit user approval. Do not auto-execute.

展示集群和内容的汇总表格、总内部链接数、预估总字数以及文件路径。在进入执行阶段前需等待用户明确确认，不得自动执行。

Strategy Import:

/blog cluster plan --from strategy [path]

策略导入：

/blog cluster plan --from strategy [path]

Bridges

blog-strategy

output into a cluster plan.

Locate strategy output. Scan the current directory (or the user-specified path) for a file containing a
```
Cluster Build Plan
```
table with the columns
```
# | Spoke Topic | Template | Target Keyword | Word Count | Internal Links
```
(the format produced by
```
/blog strategy
```
).
Parse the table. Extract the pillar row (marked
```
P
```
), the spoke rows, template assignments, target keywords, word counts, and link relationships.
Validate and enrich. Run SERP overlap validation (Plan Phase Step 2) on each keyword. Add volume estimates and verify cluster groupings semantically.
If SERP data contradicts the strategy table, flag the conflict; do not silently override the user's strategic intent.
Generate
```
cluster-plan.json
```
and
```
cluster-map.html
```
using the same outputs as the standard Plan Phase.
Present the converted plan with any SERP-based adjustments highlighted, and wait for user confirmation.

将

blog-strategy

的输出转换为集群规划方案。

定位策略输出文件。扫描当前目录（或用户指定路径），查找包含
```
Cluster Build Plan
```
表格的文件，表格需包含列
```
# | Spoke Topic | Template | Target Keyword | Word Count | Internal Links
```
（
```
/blog strategy
```
生成的格式）。
解析表格。提取支柱行（标记为
```
P
```
）、分支行、模板分配、目标关键词、字数和链接关系。
验证与丰富。对每个关键词执行SERP重叠验证（规划阶段步骤2），添加搜索量估算并验证集群分组的语义合理性。
若SERP数据与策略表格冲突，需标记冲突，不得静默覆盖用户的战略意图。
使用与标准规划阶段相同的输出格式生成
```
cluster-plan.json
```
和
```
cluster-map.html
```
。
向用户展示转换后的规划方案，并突出显示基于SERP数据的调整，等待用户确认。

Execute Phase:

/blog cluster execute [path-to-plan]

执行阶段：

/blog cluster execute [path-to-plan]

Reference:

references/execution-workflow.md

参考资料：

references/execution-workflow.md

Step 1. Load plan

步骤1：加载规划方案

Read

cluster-plan.json

from the user-specified path or the most recent

cluster-*/cluster-plan.json

in the working directory. Validate JSON structure. If no plan exists, return: "No cluster plan found. Run

/blog cluster plan <seed-keyword>

first."

从用户指定路径或工作目录中最新的

cluster-*/cluster-plan.json

读取文件。验证JSON结构。若未找到规划方案，返回：“未找到集群规划方案，请先运行

/blog cluster plan <seed-keyword>

。”

Step 2. Determine execution order

步骤2：确定执行顺序

Pillar page first (so spokes can link to a known filename).
Then spokes, ordered by
```
(cluster priority, search_volume_estimate desc, post id alphabetical)
```
. Cluster priority is the sum of estimated volumes within the cluster (highest first).
Alternating between clusters when more than 2 clusters exist diversifies the early content spread.

优先执行支柱内容（以便分支内容可以链接到已知文件名）。
然后执行分支内容，顺序为
```
(集群优先级, 搜索量估算降序, 内容ID字母顺序)
```
。集群优先级为集群内所有内容的估算搜索量之和（数值越高优先级越高）。
若集群数量超过2个，交替执行不同集群的内容，以分散早期内容分布。

Step 3. For each post: build cluster context and call

blog-write

步骤3：针对每个内容：构建集群上下文并调用

blog-write

Construct the cluster context block (full schema in

references/execution-workflow.md

) and prepend it to the topic prompt passed to the Task tool invoking

blog-write

. The context tells

blog-write

the cluster name, the post's role (pillar or spoke), the primary and secondary keywords, the chosen template, the word count target, the list of already-written posts (link to these), the list of upcoming posts (use

[INTERNAL-LINK]

placeholders), and the linking requirements for this post.

FLOW evidence triple propagation (required). The cluster context must include this directive for every spoke and the pillar: "Apply the FLOW evidence triple to every public statistic. Year anchor in prose ('In 2026,'), inline citation with publisher and title, URL with retrieval date in the source block. Drop unverifiable stats. Replace contradicted ones."

This cascade is required because cluster execution is a high-leverage operation (5 to 15 posts at once). Without explicit propagation, individual spokes could silently skip evidence discipline. See

skills/blog/references/flow-alignment.md

The context also instructs

blog-write

to run autonomously: skip topic clarification, skip outline approval, do not auto-detect template, do not pause.

Output format: standard markdown (

.md

) by default, matching

blog-write

's default. If the user explicitly requests HTML, set the platform target accordingly. Do not impose any brand-specific CSS or wordmark; that is the user's responsibility downstream.

构建集群上下文块（完整架构见

references/execution-workflow.md

），并将其添加到传递给任务工具调用

blog-write

的主题提示前。上下文信息包括集群名称、内容角色（支柱或分支）、主关键词和次要关键词、所选模板、字数目标、已完成内容列表（链接到这些内容）、待完成内容列表（使用

[INTERNAL-LINK]

占位符）以及该内容的链接要求。

FLOW证据三元组传播（必填）。集群上下文必须为每个分支内容和支柱内容包含以下指令：“对所有公开统计数据应用FLOW证据三元组。在正文中标注年份锚点（如‘2026年，’），添加包含发布方和标题的内联引用，在来源块中添加带有检索日期的URL。删除无法验证的统计数据，替换相互矛盾的数据。”

此级联要求是因为集群执行是高影响力操作（一次性生成5至15篇内容）。若没有明确的传播规则，单个分支内容可能会忽略证据规范。详见

skills/blog/references/flow-alignment.md

。

上下文还会指示

blog-write

自主运行：跳过主题澄清、跳过大纲确认、不自动检测模板、不暂停执行。

输出格式：默认使用标准Markdown（

.md

），与

blog-write

的默认格式一致。若用户明确要求HTML，则相应设置平台目标。不得添加任何特定品牌的CSS或标识，这属于用户下游的责任范围。

Step 4. Per-post optional hero image

步骤4：为单篇内容生成可选封面图

nanobanana-mcp

is configured, call

/blog image generate

via the Task tool to produce a 16:9 hero image for the post and place it in

cluster-<slug>/images/<post-slug>-hero.png

. Insert a standard markdown image reference in the post's frontmatter (

coverImage:

) and at the top of the body. If the MCP is unavailable or fails, log a warning and continue without images. Image generation is non-blocking.

若已配置

nanobanana-mcp

，则通过任务工具调用

/blog image generate

生成16:9比例的封面图，并保存到

cluster-<slug>/images/<post-slug>-hero.png

。在内容的前置元数据（

coverImage:

）和正文顶部插入标准Markdown图片引用。若MCP不可用或调用失败，则记录警告并继续执行，不生成图片。图片生成是非阻塞操作。

Step 5. Backward link injection

步骤5：反向链接注入

After each post is written:

Scan all previously written posts in the cluster directory for
```
[INTERNAL-LINK: keyword -> filename.md]
```
markers that reference the just-written post.
Replace each match with a real markdown link:
```
[keyword](filename.md)
```
.
Add a cluster metadata block to the post's frontmatter on first pass (
```
cluster:
```
,
```
cluster_role:
```
,
```
cluster_group:
```
).

每篇内容完成后：

扫描集群目录中所有已完成内容，查找引用刚完成内容的
```
[INTERNAL-LINK: keyword -> filename.md]
```
标记。
将每个匹配项替换为真实的Markdown链接：
```
[keyword](filename.md)
```
。
在首次处理时，向内容的前置元数据添加集群元数据块（
```
cluster:
```
、
```
cluster_role:
```
、
```
cluster_group:
```
）。

Step 6. Failure handling

步骤6：故障处理

blog-write

fails for a single post (timeout, error, or quality gate fail), log the failure and continue with remaining posts. Do not abort the cluster. The scorecard will mark the gap and recommend a retry with

/blog write

invoked manually for that post.

If the user cancels mid-execution, save progress and note completed posts. On the next

/blog cluster execute

, detect already-written files and resume from the next unwritten post.

若

blog-write

单篇内容执行失败（超时、错误或质量检查不通过），记录故障并继续执行剩余内容，不得中止集群执行。评分卡会标记该缺口，并建议手动调用

/blog write

重试该内容。

若用户在执行过程中取消操作，保存进度并记录已完成内容。下次调用

/blog cluster execute

时，会自动检测已完成文件并从下一个未完成内容继续执行。

Step 7. Generate

cluster-scorecard.md

步骤7：生成

cluster-scorecard.md

After all attempted posts complete, produce a markdown scorecard covering:

Per-post status (written, failed, skipped) with file path and word count.
Per-post quality score (call
```
/blog analyze
```
on each in parallel) and the cluster average.
Cluster cohesion score: a 0 to 100 composite of link reciprocity, intent diversity, template diversity, and keyword coverage (formula in
```
references/execution-workflow.md
```
).
Internal-link audit: outgoing and incoming counts per post, orphan flags, unresolved
```
[INTERNAL-LINK]
```
markers.
Cannibalization check: any two posts sharing primary keyword, or any pair with greater than 70% keyword overlap. Recommend running
```
/blog cannibalization
```
for a deeper pass.
Image generation summary: hero images generated vs. skipped.
Recommended next actions: schema generation (
```
/blog schema
```
), per-post SEO validation (
```
/blog seo-check
```
), repurposing (
```
/blog repurpose
```
).

所有尝试执行的内容完成后，生成Markdown格式的评分卡，包含以下内容：

单篇内容状态（已完成、失败、跳过），附带文件路径和字数。
单篇内容质量评分（并行调用
```
/blog analyze
```
）和集群平均评分。
集群凝聚力评分：0至100的综合评分，基于链接互惠性、意图多样性、模板多样性和关键词覆盖范围（计算公式见
```
references/execution-workflow.md
```
）。
内部链接审计：每篇内容的出站和入站链接数、孤立内容标记、未解析的
```
[INTERNAL-LINK]
```
标记。
Cannibalization检查：是否有两篇内容共享主关键词，或任意两篇内容的关键词重叠度超过70%。建议运行
```
/blog cannibalization
```
进行深度检查。
图片生成汇总：已生成封面图与跳过的数量统计。
推荐后续操作：生成Schema（
```
/blog schema
```
）、单篇内容SEO验证（
```
/blog seo-check
```
）、内容复用（
```
/blog repurpose
```
）。

Step 8. Final report

步骤8：最终报告

Return a concise summary to the user with totals, the scorecard path, and the next-action commands.

向用户返回简洁的汇总信息，包括总量统计、评分卡路径和后续操作命令。

Output Artifacts (summary)

输出产物汇总

File	Phase	Format
`cluster-plan.json`	Plan	JSON
`cluster-map.html`	Plan	Static HTML + inline SVG, no JavaScript
`pillar-<slug>.md`	Execute	Markdown (or platform-detected format)
`<spoke-slug>.md`	Execute	Markdown (or platform-detected format)
`images/<post-slug>-hero.png`	Execute (optional)	PNG via `blog-image`
`cluster-scorecard.md`	Execute	Markdown

文件	生成阶段	格式
`cluster-plan.json`	规划阶段	JSON
`cluster-map.html`	规划阶段	静态HTML + 内联SVG，无JavaScript
`pillar-<slug>.md`	执行阶段	Markdown（或平台检测格式）
`<spoke-slug>.md`	执行阶段	Markdown（或平台检测格式）
`images/<post-slug>-hero.png`	执行阶段（可选）	PNG格式，由 `blog-image` 生成
`cluster-scorecard.md`	执行阶段	Markdown

Quality Gates

质量检查门限

Gate	Check	Action on fail
Cluster minimum	At least 2 clusters with at least 2 posts each	Warn during plan; suggest expansion
Cannibalization	No two posts share primary keyword	Block execution; require plan adjustment
Link completeness	Every post has 3 or more incoming internal links	Warn in scorecard
Word count	Pillar at least 2,500 words; spokes at least 1,200 words	Pass to `blog-write` as a hard constraint
Intent diversity	At least 2 distinct intents across clusters	Warn in scorecard
Template diversity	At least 3 distinct templates across the cluster	Warn in scorecard

检查项	检查内容	失败处理动作
集群最小规模	至少2个集群，每个集群至少包含2篇内容	规划阶段发出警告；建议扩展集群
关键词Cannibalization	无两篇内容共享主关键词	阻止执行；要求调整规划方案
链接完整性	每篇内容至少有3个入站内部链接	在评分卡中发出警告
字数要求	支柱内容至少2500字；分支内容至少1200字	作为硬约束传递给 `blog-write`
意图多样性	集群间至少包含2种不同的意图类型	在评分卡中发出警告
模板多样性	集群内至少包含3种不同的模板类型	在评分卡中发出警告

Error Handling

错误处理

Scenario	Action
Seed keyword too broad (more than 50 keyword variants)	Suggest narrowing the focus before clustering.
Seed keyword too narrow (fewer than 5 keyword variants)	Offer a smaller cluster (pillar plus 2 to 3 spokes) or suggest broadening.
WebSearch unavailable	Fall back to Claude's reasoning for keyword expansion and grouping. Note the reduced accuracy in the scorecard.
`blog-write` fails for one post	Log, skip, continue. Mark the gap in the scorecard.
`blog-write` not installed	Return: "blog-cluster requires claude-blog. Install it before running this skill."
`cluster-plan.json` malformed	Validate JSON and report parse errors with line numbers.
User cancels execution	Save progress; resume on next invocation with already-written posts auto-detected.
`nanobanana-mcp` not configured	Skip hero image generation; warn once at start of execute, not per post.

场景	处理动作
种子关键词过于宽泛（超过50个关键词变体）	建议缩小聚焦范围后再进行聚类。
种子关键词过于狭窄（少于5个关键词变体）	提供小型集群方案（支柱内容+2至3篇分支内容）或建议扩大关键词范围。
WebSearch不可用	退回到Claude的推理能力进行关键词扩展和分组。在评分卡中注明准确性降低。
`blog-write` 单篇内容执行失败	记录故障、跳过该内容、继续执行。在评分卡中标记该缺口。
`blog-write` 未安装	返回：“blog-cluster依赖claude-blog，请先安装后再运行本技能。”
`cluster-plan.json` 格式错误	验证JSON并报告解析错误及行号。
用户取消执行	保存进度；下次调用时自动检测已完成内容并继续执行。
`nanobanana-mcp` 未配置	跳过封面图生成；执行开始时发出一次警告，而非每篇内容都警告。

Differentiation from related claude-blog skills

与同类claude-blog技能的差异

Skill	Role	What blog-cluster adds
`blog-strategy`	Plans 3 to 5 content pillars and draws hub-and-spoke diagrams as a strategic exercise	Performs SERP-based semantic clustering, then executes the plan into real, interlinked posts.
`blog-calendar`	Schedules publication dates around topic clusters	Does not build clusters or write posts; this skill does both.
`blog-cannibalization`	Detects keyword overlap in existing content	Diagnostic only. blog-cluster prevents cannibalization at the planning stage.
`blog-write`	Writes one post at a time	blog-cluster orchestrates many `blog-write` calls with shared cluster context and bidirectional linking.
`blog-outline`	Generates a single SERP-informed outline	blog-cluster generates an outline-equivalent across an entire cluster, then writes the posts.

blog-cluster is the general contractor: it analyzes the topic, draws the data-driven plan, and builds the entire structure from a single seed keyword.

技能	角色	blog-cluster新增功能
`blog-strategy`	规划3至5个内容支柱并绘制Hub-and-Spoke图作为战略规划	基于SERP进行语义聚类，然后将规划方案执行成真实的互联内容。
`blog-calendar`	围绕主题集群安排发布日期	不构建集群或撰写内容；本技能兼具这两项功能。
`blog-cannibalization`	检测现有内容中的关键词重叠问题	仅用于诊断。blog-cluster在规划阶段就防止关键词Cannibalization。
`blog-write`	一次撰写一篇内容	blog-cluster协调多次 `blog-write` 调用，附带共享集群上下文并实现双向链接。
`blog-outline`	生成单个基于SERP的大纲	blog-cluster为整个集群生成等效的大纲，然后撰写所有内容。

blog-cluster相当于总承包商：它分析主题、制定数据驱动的规划方案，并从单个种子关键词搭建完整的内容体系。

blog-cluster

Original

Translation

Blog Cluster (Semantic Topic Cluster Engine)

博客集群（语义主题集群引擎）

Commands

命令

Key references (load on demand)

关键参考资料（按需加载）

Cross-references to existing claude-blog skills

与现有claude-blog技能的交叉引用

Command Routing

命令路由逻辑

Plan Phase: /blog cluster plan <seed-keyword>

规划阶段：/blog cluster plan <seed-keyword>

Step 1. Seed keyword expansion

步骤1：种子关键词扩展

Step 2. Semantic clustering

步骤2：语义聚类

Step 3. Cluster architecture design

步骤3：集群架构设计

Step 4. Internal link matrix

步骤4：内部链接矩阵

Step 5. Generate output files

步骤5：生成输出文件

cluster-plan.json schema

cluster-plan.json 架构

cluster-map.html (XSS-safe)

cluster-map.html（防XSS安全）

Step 6. Present plan to user

步骤6：向用户展示规划方案

Strategy Import: /blog cluster plan --from strategy [path]

策略导入：/blog cluster plan --from strategy [path]

Execute Phase: /blog cluster execute [path-to-plan]

执行阶段：/blog cluster execute [path-to-plan]

Step 1. Load plan

步骤1：加载规划方案

Step 2. Determine execution order

步骤2：确定执行顺序

Step 3. For each post: build cluster context and call blog-write

步骤3：针对每个内容：构建集群上下文并调用blog-write

Step 4. Per-post optional hero image

步骤4：为单篇内容生成可选封面图

Step 5. Backward link injection

步骤5：反向链接注入

Step 6. Failure handling

步骤6：故障处理

Step 7. Generate cluster-scorecard.md

步骤7：生成cluster-scorecard.md

Step 8. Final report

步骤8：最终报告

Output Artifacts (summary)

输出产物汇总

Quality Gates

质量检查门限

Error Handling

错误处理

Differentiation from related claude-blog skills

与同类claude-blog技能的差异

Plan Phase:
`/blog cluster plan <seed-keyword>`

规划阶段：
`/blog cluster plan <seed-keyword>`

`cluster-plan.json`
schema

`cluster-plan.json`
架构

`cluster-map.html`
(XSS-safe)

`cluster-map.html`
（防XSS安全）

Strategy Import:
`/blog cluster plan --from strategy [path]`

策略导入：
`/blog cluster plan --from strategy [path]`

Execute Phase:
`/blog cluster execute [path-to-plan]`

执行阶段：
`/blog cluster execute [path-to-plan]`

Step 3. For each post: build cluster context and call
`blog-write`

步骤3：针对每个内容：构建集群上下文并调用
`blog-write`

Step 7. Generate
`cluster-scorecard.md`

步骤7：生成
`cluster-scorecard.md`