[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7ENGD70WMXWQW9RQMV5ZYC4":3,"skill-same:01M7ENGD70WMXWQW9RQMV5ZYC4":67},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":32,"commitSha":33,"skillDirectory":34,"forksCount":35,"repositoryHtmlUrl":36,"repositoryHomepageUrl":6,"ownerHtmlUrl":37,"ownerBlogUrl":38,"ownerDescription":6,"updatedAt":39,"install":40,"resources":52},"01M7ENGD70WMXWQW9RQMV5ZYC4","blog-cluster",null,"Semantic topic cluster planning and automated execution engine for claude-blog. Performs SERP-based keyword research, groups keywords by search intent and SERP overlap, builds a hub-and-spoke cluster architecture, generates an interactive SVG cluster map, and executes the full cluster by orchestrating blog-write calls with shared cluster context and automatic internal-link injection. Fills the strategy-to-execution gap: blog-strategy plans the blueprint, blog-cluster builds the house. Use when user says \"blog cluster\", \"topic cluster\", \"content cluster\", \"cluster plan\", \"cluster execute\", \"pillar content\", \"hub and spoke\", \"content ecosystem\", \"cluster map\".","TanNhatCMS\u002Fjarvis","TanNhatCMS","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F113796420?v=4","01M753MV17KG5CD3152G3WQEQY","KietNT","01M7EKEDYVYQ6PWVJJMZ911VMX",0,"MIT","https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis\u002Fblob\u002Fmain\u002Fskills\u002Fams-imp-blog-cluster\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: blog-cluster\ndescription: >\n  Semantic topic cluster planning and automated execution engine for claude-blog.\n  Performs SERP-based keyword research, groups keywords by search intent and\n  SERP overlap, builds a hub-and-spoke cluster architecture, generates an\n  interactive SVG cluster map, and executes the full cluster by orchestrating\n  blog-write calls with shared cluster context and automatic internal-link\n  injection. Fills the strategy-to-execution gap: blog-strategy plans the\n  blueprint, blog-cluster builds the house.\n  Use when user says \"blog cluster\", \"topic cluster\", \"content cluster\",\n  \"cluster plan\", \"cluster execute\", \"pillar content\", \"hub and spoke\",\n  \"content ecosystem\", \"cluster map\".\nlicense: MIT\ncompatibility: Requires Claude Code and claude-blog (provides blog-write, blog-chart, blog-image)\nmetadata:\n  author: AgriciDaniel\n  version: \"2.2.0\"\n  category: blog\nuser-invokable: true\nargument-hint: \"[plan|execute] [seed-keyword|cluster-plan.json]\"\n---\n\n# Blog Cluster (Semantic Topic Cluster Engine)\n\nPlans and executes entire interlinked content ecosystems from a single seed\nkeyword. Three layers: Semantic Clustering (the brain), Cluster Architecture\n(the structure), and Execution Engine (the machine).\n\n> Adapted from the **semantic-cluster-engine** submission by Lutfiya Miller\n> (winner, AI Marketing Hub Pro Challenge, March 2026, 95\u002F100 Exemplary).\n> Original repository: https:\u002F\u002Fgithub.com\u002FDrfiya\u002Fsemantic-cluster-engine\n> This port keeps the Plan + Execute architecture and the cluster context\n> innovation, removes brand-specific (ScienceExperts.ai) styling and image\n> prompts, and routes through claude-blog's existing sub-skills.\n\n## Commands\n\n| Command | What it does |\n|---------|--------------|\n| `\u002Fblog cluster` | Interactive. Asks whether to plan or execute. |\n| `\u002Fblog cluster plan \u003Cseed-keyword>` | SERP-based semantic analysis. Outputs cluster plan + map. |\n| `\u002Fblog cluster plan --from strategy [path]` | Imports existing `blog-strategy` cluster build plan and validates against SERP data. |\n| `\u002Fblog cluster execute [path-to-plan]` | Sequential `blog-write` calls with cluster context and auto-interlinks. |\n\n## Key references (load on demand)\n\n- `references\u002Fsemantic-clustering.md` (SERP overlap analysis, intent classification, keyword universe expansion)\n- `references\u002Fcluster-architecture.md` (hub-and-spoke specs, schema strategy, link-density rules)\n- `references\u002Fexecution-workflow.md` (execution order, context injection, scorecard, failure handling)\n\n## Cross-references to existing claude-blog skills\n\n| Skill | When this skill calls it |\n|-------|--------------------------|\n| `\u002Fblog strategy` | Upstream planning. `plan --from strategy` consumes its Cluster Build Plan tables. |\n| `\u002Fblog write` | Per-post execution. Each spoke and the pillar are produced by `blog-write` with a prepended cluster-context block. |\n| `\u002Fblog chart` | Invoked internally by `blog-write` for inline SVG charts. No direct call from this skill. |\n| `\u002Fblog image` | Optional hero image generation per post. Model selection is delegated to `blog-image`; prefer current Gemini image models when available. |\n| `\u002Fblog seo-check` | Recommended after execution for per-post on-page validation. |\n| `\u002Fblog cannibalization` | Recommended after execution to confirm zero keyword overlap across the cluster. |\n| `\u002Fblog schema` | Recommended after execution to add `BreadcrumbList`, `ItemList`, and `Article` schema. |\n\nThis skill never modifies files belonging to other skills. It calls them via the Task tool or as orchestrated sub-skills.\n\n## Command Routing\n\n1. Parse the user's command to determine the sub-command.\n2. If the user typed only `\u002Fblog cluster`, ask: \"Would you like to **plan** a new cluster or **execute** an existing plan?\"\n3. Route:\n   - `plan \u003Ckeyword>` to the Plan Phase (below)\n   - `plan --from strategy [path]` to the Strategy Import flow (below)\n   - `execute [path]`, `build`, or `run` to the Execute Phase (below)\n\n---\n\n## Plan Phase: `\u002Fblog cluster plan \u003Cseed-keyword>`\n\nReference: `references\u002Fsemantic-clustering.md`\n\n### Step 1. Seed keyword expansion\n\nUse WebSearch to expand the seed into a keyword universe of 30 to 50 phrases:\n\n1. Direct search of `\u003Cseed>` to capture related searches and \"People also ask\".\n2. Long-tail expansion: `\u003Cseed> guide`, `\u003Cseed> tips`, `\u003Cseed> tools`, `\u003Cseed> examples`, `\u003Cseed> vs`, `best \u003Cseed>`, `how to \u003Cseed>`.\n3. Question mining: `what is \u003Cseed>`, `how does \u003Cseed> work`, `why \u003Cseed>`, `\u003Cseed> for beginners`.\n4. Intent variants: add commercial modifiers (best, top, review, comparison, pricing), informational modifiers (guide, tutorial, explained, examples), and transactional modifiers (buy, download, tool, software, service).\n5. Year freshness: `\u003Cseed> 2026`.\n\n### Step 2. Semantic clustering\n\nGroup the expanded keywords using the priority rules in `references\u002Fsemantic-clustering.md`:\n\n1. **SERP Overlap Analysis** is the primary signal. Two keywords with 4 or more shared top-10 results usually target the same intent and should be considered for one post.\n2. **Intent Classification** assigns each keyword to informational, commercial, transactional, or navigational.\n3. **Entity Mapping** identifies the people, products, frameworks, and organizations Google associates with the topic.\n4. **Grouping** combines keywords that share intent and topical proximity. Each group becomes one branch of the hub and spoke.\n\n### Step 3. Cluster architecture design\n\nReference: `references\u002Fcluster-architecture.md`\n\nBuild the hub and spoke:\n\n- **Pillar (hub)**: targets the broadest keyword. Word count 2,500 to 4,000. Template `pillar-page`. Links down to every spoke.\n- **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.\n\nCluster formation rules:\n\n- Normal mode: 2 to 5 clusters per pillar, 2 to 4 spokes per cluster, total 1 pillar plus 5 to 15 spokes.\n- Small-cluster mode: for narrow seeds, allow 1 pillar plus 2 to 4 spokes only after warning the user and asking for confirmation.\n- Every spoke targets a unique primary keyword (zero cannibalization).\n\n### Step 4. Internal link matrix\n\nFor each spoke `S`:\n\n- `S` to Pillar (always; anchor text uses the pillar's primary keyword).\n- Pillar to `S` (always; anchor text uses `S`'s primary keyword).\n- `S` to other spokes in the same cluster (2 to 3 links each, contextual anchors).\n- `S` to spokes in adjacent clusters (0 to 1 links, only when semantically relevant).\n\nVerify every spoke has at least 2 incoming links. Count total planned interlinks.\n\n### Step 5. Generate output files\n\nAll plan and execute artifacts go into a single subdirectory of the current working directory. Canonicalize the output directory, refuse symlinks, and reject writes outside the cluster directory. Slugs must match lowercase letters, numbers, and hyphens only; reject absolute paths, `..`, path separators inside slugs, and hidden control characters.\n\n```\n\u003Ccwd>\u002F\n└── cluster-\u003Cseed-keyword-slug>\u002F\n    ├── cluster-plan.json\n    ├── cluster-map.html\n    ├── pillar-\u003Cslug>.md       (Execute Phase)\n    ├── \u003Cspoke-slug>.md        (Execute Phase, one per spoke)\n    └── cluster-scorecard.md   (Execute Phase)\n```\n\n#### `cluster-plan.json` schema\n\n```json\n{\n  \"seed_keyword\": \"\u003Cseed>\",\n  \"generated_at\": \"YYYY-MM-DDTHH:MM:SSZ\",\n  \"pillar\": {\n    \"id\": \"P\",\n    \"title\": \"Title of the pillar\",\n    \"primary_keyword\": \"broadest keyword\",\n    \"secondary_keywords\": [\"...\"],\n    \"search_volume_estimate\": \"high|medium|low\",\n    \"template\": \"pillar-page\",\n    \"word_count_target\": 3000,\n    \"cluster\": \"pillar\"\n  },\n  \"clusters\": [\n    {\n      \"name\": \"Cluster A: Theme\",\n      \"intent\": \"informational|commercial|transactional\",\n      \"color\": \"#2563eb\",\n      \"posts\": [\n        {\n          \"id\": \"A1\",\n          \"title\": \"Post title\",\n          \"primary_keyword\": \"long-tail keyword\",\n          \"secondary_keywords\": [\"...\"],\n          \"search_volume_estimate\": \"high|medium|low\",\n          \"template\": \"how-to-guide\",\n          \"word_count_target\": 1500,\n          \"links_to\": [\"P\", \"A2\"],\n          \"links_from\": [\"P\", \"A2\"]\n        }\n      ]\n    }\n  ],\n  \"total_posts\": 9,\n  \"total_interlinks\": 23,\n  \"estimated_total_words\": 18000\n}\n```\n\nNote: 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).\n\n#### `cluster-map.html` (XSS-safe)\n\nA static, self-contained HTML file with an embedded SVG visualization. Hard rules for the writer:\n\n- No inline `\u003Cscript>` blocks. No `onclick`, `onmouseover`, or any `on*` event attributes anywhere in the document.\n- No external script `\u003Csrc>` references.\n- Every text label drawn into the SVG (titles, keywords, cluster names) must be escaped: replace `&` with `&amp;`, `\u003C` with `&lt;`, `>` with `&gt;`, `\"` with `&quot;`, and `'` with `&#39;` before insertion.\n- Hover effects use CSS `:hover` only. No JavaScript.\n- Use `\u003Ctitle>` child elements inside SVG nodes for accessible tooltips (browser native, no script).\n\nThe map shows: a central pillar node, color-coded cluster groups radiating outward, spoke nodes within each cluster, and link lines connecting related nodes.\n\n### Step 6. Present plan to user\n\nShow 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.\n\n---\n\n## Strategy Import: `\u002Fblog cluster plan --from strategy [path]`\n\nBridges `blog-strategy` output into a cluster plan.\n\n1. 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 `\u002Fblog strategy`).\n2. Parse the table. Extract the pillar row (marked `P`), the spoke rows, template assignments, target keywords, word counts, and link relationships.\n3. Validate and enrich. Run SERP overlap validation (Plan Phase Step 2) on each keyword. Add volume estimates and verify cluster groupings semantically.\n4. If SERP data contradicts the strategy table, flag the conflict; do not silently override the user's strategic intent.\n5. Generate `cluster-plan.json` and `cluster-map.html` using the same outputs as the standard Plan Phase.\n6. Present the converted plan with any SERP-based adjustments highlighted, and wait for user confirmation.\n\n---\n\n## Execute Phase: `\u002Fblog cluster execute [path-to-plan]`\n\nReference: `references\u002Fexecution-workflow.md`\n\n### Step 1. Load plan\n\nRead `cluster-plan.json` from the user-specified path or the most recent `cluster-*\u002Fcluster-plan.json` in the working directory. Validate JSON structure. If no plan exists, return: \"No cluster plan found. Run `\u002Fblog cluster plan \u003Cseed-keyword>` first.\"\n\nBefore reading a user-supplied plan path, canonicalize it relative to the current working directory. Reject absolute paths, `..`, symlinks, non-`cluster-plan.json` filenames, and any path outside the selected cluster directory. Constrain all generated post, image, map, and scorecard outputs to that cluster directory.\n\n### Step 2. Determine execution order\n\n1. Pillar page first (so spokes can link to a known filename).\n2. 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).\n3. Alternating between clusters when more than 2 clusters exist diversifies the early content spread.\n\n### Step 3. For each post: build cluster context and call `blog-write`\n\nConstruct the cluster context block (full schema in `references\u002Fexecution-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.\n\n**Evidence provenance propagation.** The cluster context includes this directive\nfor every spoke and the pillar: \"Keep material claims traceable to sources that\nsupport them. Record dates, publisher\u002Ftitle details, retrieval notes,\nmethodology, and limitations when they help identify or interpret the source.\nUse the publication's citation style. Drop unverifiable statistics and replace\ncontradicted ones.\"\n\nThis cascade preserves evidence discipline across batch execution without\nturning a fixed source-record format into a score or gate. See\n`skills\u002Fblog\u002Freferences\u002Fflow-alignment.md`.\n\nThe context also instructs `blog-write` to run autonomously: skip topic clarification, skip outline approval, do not auto-detect template, do not pause.\n\nOutput 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.\n\n### Step 4. Per-post optional hero image\n\nIf `blog-image` is available, call `\u002Fblog image generate` via the Task tool to produce a 16:9 hero image for the post and place it in `cluster-\u003Cslug>\u002Fimages\u002F\u003Cpost-slug>-hero.png`. Delegate provider selection to `blog-image`, prefer current Gemini image models when available, and record the model ID in the scorecard. If image generation is unavailable or fails, log a warning and continue without images. Image generation is non-blocking.\n\n### Step 5. Backward link injection\n\nAfter each post is written:\n\n1. Scan all previously written posts in the cluster directory for `[INTERNAL-LINK: keyword -> filename.md]` markers that reference the just-written post.\n2. Replace each match with a real markdown link: `[keyword](filename.md)`.\n3. Add a cluster metadata block to the post's frontmatter on first pass (`cluster:`, `cluster_role:`, `cluster_group:`).\n\n### Step 6. Failure handling\n\nIf `blog-write` returns a quality-gate failure for any post, stop the batch immediately. Save progress, mark the failed post and all remaining posts as skipped, and tell the user to inspect or retry the failed post manually before continuing. Do not keep generating 5 to 15 posts after a quality failure, because that can create scaled-content-abuse risk.\n\nIf `blog-write` fails before content is generated because of a timeout or runtime error, log the failure in the scorecard and stop unless the user explicitly resumes after checking the cause.\n\nIf the user cancels mid-execution, save progress and note completed posts. On the next `\u002Fblog cluster execute`, detect already-written files and resume from the next unwritten post.\n\n### Step 7. Generate `cluster-scorecard.md`\n\nAfter all attempted posts complete, produce a markdown scorecard covering:\n\n- Per-post status (written, failed, skipped) with file path and word count.\n- Per-post quality score (call `\u002Fblog analyze` on each in parallel via Task) and the cluster average.\n- Cluster cohesion score: a 0 to 100 composite of link reciprocity, intent diversity, template diversity, and keyword coverage (formula in `references\u002Fexecution-workflow.md`).\n- Internal-link audit: outgoing and incoming counts per post, orphan flags, unresolved `[INTERNAL-LINK]` markers.\n- Cannibalization check: any two posts sharing primary keyword, or any pair with greater than 70% keyword overlap. Recommend running `\u002Fblog cannibalization` for a deeper pass.\n- Image generation summary: hero images generated vs. skipped.\n- Recommended next actions: schema generation (`\u002Fblog schema`), per-post SEO validation (`\u002Fblog seo-check`), repurposing (`\u002Fblog repurpose`).\n\n### Step 8. Final report\n\nReturn a concise summary to the user with totals, the scorecard path, and the next-action commands.\n\n---\n\n## Output Artifacts (summary)\n\n| File | Phase | Format |\n|------|-------|--------|\n| `cluster-plan.json` | Plan | JSON |\n| `cluster-map.html` | Plan | Static HTML + inline SVG, no JavaScript |\n| `pillar-\u003Cslug>.md` | Execute | Markdown (or platform-detected format) |\n| `\u003Cspoke-slug>.md` | Execute | Markdown (or platform-detected format) |\n| `images\u002F\u003Cpost-slug>-hero.png` | Execute (optional) | PNG via `blog-image` |\n| `cluster-scorecard.md` | Execute | Markdown |\n\n---\n\n## Quality Gates\n\n| Gate | Check | Action on fail |\n|------|-------|----------------|\n| Cluster minimum | At least 2 clusters with at least 2 posts each | Warn during plan; suggest expansion |\n| Cannibalization | No two posts share primary keyword | Block execution; require plan adjustment |\n| Link completeness | Every post has 2 or more incoming internal links | Warn in scorecard |\n| Length estimate | Pillars often need more depth than spokes | Pass to `blog-write` as optional planning context; never enforce a raw word minimum |\n| Intent diversity | At least 2 distinct intents across clusters | Warn in scorecard |\n| Template diversity | At least 3 distinct templates across the cluster | Warn in scorecard |\n\n---\n\n## Error Handling\n\n| Scenario | Action |\n|----------|--------|\n| Seed keyword too broad (more than 50 keyword variants) | Suggest narrowing the focus before clustering. |\n| Seed keyword too narrow (fewer than 5 keyword variants) | Offer a smaller cluster (pillar plus 2 to 3 spokes) or suggest broadening. |\n| WebSearch unavailable | Produce a draft plan only and require user confirmation, imported keyword data, or SERP data before execution. |\n| `blog-write` fails a quality gate or content generation | Stop the batch, save progress, mark remaining posts as skipped, and require manual inspection or explicit resume. |\n| `blog-write` not installed | Return: \"blog-cluster requires claude-blog. Install it before running this skill.\" |\n| `cluster-plan.json` malformed | Validate JSON and report parse errors with line numbers. |\n| User cancels execution | Save progress; resume on next invocation with already-written posts auto-detected. |\n| Optional image generation unavailable or failed | Skip hero image generation; warn once at start of execute, not per post. |\n\n---\n\n## Differentiation from related claude-blog skills\n\n| Skill | Role | What blog-cluster adds |\n|-------|------|------------------------|\n| `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. |\n| `blog-calendar` | Schedules publication dates around topic clusters | Does not build clusters or write posts; this skill does both. |\n| `blog-cannibalization` | Detects keyword overlap in existing content | Diagnostic only. blog-cluster prevents cannibalization at the planning stage. |\n| `blog-write` | Writes one post at a time | blog-cluster orchestrates many `blog-write` calls with shared cluster context and bidirectional linking. |\n| `blog-outline` | Generates a single SERP-informed outline | blog-cluster generates an outline-equivalent across an entire cluster, then writes the posts. |\n\nblog-cluster is the general contractor: it analyzes the topic, draws the data-driven plan, and builds the entire structure from a single seed keyword.\n","# Blog Cluster (Semantic Topic Cluster Engine)\n\nPlans and executes entire interlinked content ecosystems from a single seed\nkeyword. Three layers: Semantic Clustering (the brain), Cluster Architecture\n(the structure), and Execution Engine (the machine).\n\n> Adapted from the **semantic-cluster-engine** submission by Lutfiya Miller\n> (winner, AI Marketing Hub Pro Challenge, March 2026, 95\u002F100 Exemplary).\n> Original repository: https:\u002F\u002Fgithub.com\u002FDrfiya\u002Fsemantic-cluster-engine\n> This port keeps the Plan + Execute architecture and the cluster context\n> innovation, removes brand-specific (ScienceExperts.ai) styling and image\n> prompts, and routes through claude-blog's existing sub-skills.\n\n## Commands\n\n| Command | What it does |\n|---------|--------------|\n| `\u002Fblog cluster` | Interactive. Asks whether to plan or execute. |\n| `\u002Fblog cluster plan \u003Cseed-keyword>` | SERP-based semantic analysis. Outputs cluster plan + map. |\n| `\u002Fblog cluster plan --from strategy [path]` | Imports existing `blog-strategy` cluster build plan and validates against SERP data. |\n| `\u002Fblog cluster execute [path-to-plan]` | Sequential `blog-write` calls with cluster context and auto-interlinks. |\n\n## Key references (load on demand)\n\n- `references\u002Fsemantic-clustering.md` (SERP overlap analysis, intent classification, keyword universe expansion)\n- `references\u002Fcluster-architecture.md` (hub-and-spoke specs, schema strategy, link-density rules)\n- `references\u002Fexecution-workflow.md` (execution order, context injection, scorecard, failure handling)\n\n## Cross-references to existing claude-blog skills\n\n| Skill | When this skill calls it |\n|-------|--------------------------|\n| `\u002Fblog strategy` | Upstream planning. `plan --from strategy` consumes its Cluster Build Plan tables. |\n| `\u002Fblog write` | Per-post execution. Each spoke and the pillar are produced by `blog-write` with a prepended cluster-context block. |\n| `\u002Fblog chart` | Invoked internally by `blog-write` for inline SVG charts. No direct call from this skill. |\n| `\u002Fblog image` | Optional hero image generation per post. Model selection is delegated to `blog-image`; prefer current Gemini image models when available. |\n| `\u002Fblog seo-check` | Recommended after execution for per-post on-page validation. |\n| `\u002Fblog cannibalization` | Recommended after execution to confirm zero keyword overlap across the cluster. |\n| `\u002Fblog schema` | Recommended after execution to add `BreadcrumbList`, `ItemList`, and `Article` schema. |\n\nThis skill never modifies files belonging to other skills. It calls them via the Task tool or as orchestrated sub-skills.\n\n## Command Routing\n\n1. Parse the user's command to determine the sub-command.\n2. If the user typed only `\u002Fblog cluster`, ask: \"Would you like to **plan** a new cluster or **execute** an existing plan?\"\n3. Route:\n   - `plan \u003Ckeyword>` to the Plan Phase (below)\n   - `plan --from strategy [path]` to the Strategy Import flow (below)\n   - `execute [path]`, `build`, or `run` to the Execute Phase (below)\n\n---\n\n## Plan Phase: `\u002Fblog cluster plan \u003Cseed-keyword>`\n\nReference: `references\u002Fsemantic-clustering.md`\n\n### Step 1. Seed keyword expansion\n\nUse WebSearch to expand the seed into a keyword universe of 30 to 50 phrases:\n\n1. Direct search of `\u003Cseed>` to capture related searches and \"People also ask\".\n2. Long-tail expansion: `\u003Cseed> guide`, `\u003Cseed> tips`, `\u003Cseed> tools`, `\u003Cseed> examples`, `\u003Cseed> vs`, `best \u003Cseed>`, `how to \u003Cseed>`.\n3. Question mining: `what is \u003Cseed>`, `how does \u003Cseed> work`, `why \u003Cseed>`, `\u003Cseed> for beginners`.\n4. Intent variants: add commercial modifiers (best, top, review, comparison, pricing), informational modifiers (guide, tutorial, explained, examples), and transactional modifiers (buy, download, tool, software, service).\n5. Year freshness: `\u003Cseed> 2026`.\n\n### Step 2. Semantic clustering\n\nGroup the expanded keywords using the priority rules in `references\u002Fsemantic-clustering.md`:\n\n1. **SERP Overlap Analysis** is the primary signal. Two keywords with 4 or more shared top-10 results usually target the same intent and should be considered for one post.\n2. **Intent Classification** assigns each keyword to informational, commercial, transactional, or navigational.\n3. **Entity Mapping** identifies the people, products, frameworks, and organizations Google associates with the topic.\n4. **Grouping** combines keywords that share intent and topical proximity. Each group becomes one branch of the hub and spoke.\n\n### Step 3. Cluster architecture design\n\nReference: `references\u002Fcluster-architecture.md`\n\nBuild the hub and spoke:\n\n- **Pillar (hub)**: targets the broadest keyword. Word count 2,500 to 4,000. Template `pillar-page`. Links down to every spoke.\n- **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.\n\nCluster formation rules:\n\n- Normal mode: 2 to 5 clusters per pillar, 2 to 4 spokes per cluster, total 1 pillar plus 5 to 15 spokes.\n- Small-cluster mode: for narrow seeds, allow 1 pillar plus 2 to 4 spokes only after warning the user and asking for confirmation.\n- Every spoke targets a unique primary keyword (zero cannibalization).\n\n### Step 4. Internal link matrix\n\nFor each spoke `S`:\n\n- `S` to Pillar (always; anchor text uses the pillar's primary keyword).\n- Pillar to `S` (always; anchor text uses `S`'s primary keyword).\n- `S` to other spokes in the same cluster (2 to 3 links each, contextual anchors).\n- `S` to spokes in adjacent clusters (0 to 1 links, only when semantically relevant).\n\nVerify every spoke has at least 2 incoming links. Count total planned interlinks.\n\n### Step 5. Generate output files\n\nAll plan and execute artifacts go into a single subdirectory of the current working directory. Canonicalize the output directory, refuse symlinks, and reject writes outside the cluster directory. Slugs must match lowercase letters, numbers, and hyphens only; reject absolute paths, `..`, path separators inside slugs, and hidden control characters.\n\n```\n\u003Ccwd>\u002F\n└── cluster-\u003Cseed-keyword-slug>\u002F\n    ├── cluster-plan.json\n    ├── cluster-map.html\n    ├── pillar-\u003Cslug>.md       (Execute Phase)\n    ├── \u003Cspoke-slug>.md        (Execute Phase, one per spoke)\n    └── cluster-scorecard.md   (Execute Phase)\n```\n\n#### `cluster-plan.json` schema\n\n```json\n{\n  \"seed_keyword\": \"\u003Cseed>\",\n  \"generated_at\": \"YYYY-MM-DDTHH:MM:SSZ\",\n  \"pillar\": {\n    \"id\": \"P\",\n    \"title\": \"Title of the pillar\",\n    \"primary_keyword\": \"broadest keyword\",\n    \"secondary_keywords\": [\"...\"],\n    \"search_volume_estimate\": \"high|medium|low\",\n    \"template\": \"pillar-page\",\n    \"word_count_target\": 3000,\n    \"cluster\": \"pillar\"\n  },\n  \"clusters\": [\n    {\n      \"name\": \"Cluster A: Theme\",\n      \"intent\": \"informational|commercial|transactional\",\n      \"color\": \"#2563eb\",\n      \"posts\": [\n        {\n          \"id\": \"A1\",\n          \"title\": \"Post title\",\n          \"primary_keyword\": \"long-tail keyword\",\n          \"secondary_keywords\": [\"...\"],\n          \"search_volume_estimate\": \"high|medium|low\",\n          \"template\": \"how-to-guide\",\n          \"word_count_target\": 1500,\n          \"links_to\": [\"P\", \"A2\"],\n          \"links_from\": [\"P\", \"A2\"]\n        }\n      ]\n    }\n  ],\n  \"total_posts\": 9,\n  \"total_interlinks\": 23,\n  \"estimated_total_words\": 18000\n}\n```\n\nNote: 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).\n\n#### `cluster-map.html` (XSS-safe)\n\nA static, self-contained HTML file with an embedded SVG visualization. Hard rules for the writer:\n\n- No inline `\u003Cscript>` blocks. No `onclick`, `onmouseover`, or any `on*` event attributes anywhere in the document.\n- No external script `\u003Csrc>` references.\n- Every text label drawn into the SVG (titles, keywords, cluster names) must be escaped: replace `&` with `&amp;`, `\u003C` with `&lt;`, `>` with `&gt;`, `\"` with `&quot;`, and `'` with `&#39;` before insertion.\n- Hover effects use CSS `:hover` only. No JavaScript.\n- Use `\u003Ctitle>` child elements inside SVG nodes for accessible tooltips (browser native, no script).\n\nThe map shows: a central pillar node, color-coded cluster groups radiating outward, spoke nodes within each cluster, and link lines connecting related nodes.\n\n### Step 6. Present plan to user\n\nShow 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.\n\n---\n\n## Strategy Import: `\u002Fblog cluster plan --from strategy [path]`\n\nBridges `blog-strategy` output into a cluster plan.\n\n1. 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 `\u002Fblog strategy`).\n2. Parse the table. Extract the pillar row (marked `P`), the spoke rows, template assignments, target keywords, word counts, and link relationships.\n3. Validate and enrich. Run SERP overlap validation (Plan Phase Step 2) on each keyword. Add volume estimates and verify cluster groupings semantically.\n4. If SERP data contradicts the strategy table, flag the conflict; do not silently override the user's strategic intent.\n5. Generate `cluster-plan.json` and `cluster-map.html` using the same outputs as the standard Plan Phase.\n6. Present the converted plan with any SERP-based adjustments highlighted, and wait for user confirmation.\n\n---\n\n## Execute Phase: `\u002Fblog cluster execute [path-to-plan]`\n\nReference: `references\u002Fexecution-workflow.md`\n\n### Step 1. Load plan\n\nRead `cluster-plan.json` from the user-specified path or the most recent `cluster-*\u002Fcluster-plan.json` in the working directory. Validate JSON structure. If no plan exists, return: \"No cluster plan found. Run `\u002Fblog cluster plan \u003Cseed-keyword>` first.\"\n\nBefore reading a user-supplied plan path, canonicalize it relative to the current working directory. Reject absolute paths, `..`, symlinks, non-`cluster-plan.json` filenames, and any path outside the selected cluster directory. Constrain all generated post, image, map, and scorecard outputs to that cluster directory.\n\n### Step 2. Determine execution order\n\n1. Pillar page first (so spokes can link to a known filename).\n2. 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).\n3. Alternating between clusters when more than 2 clusters exist diversifies the early content spread.\n\n### Step 3. For each post: build cluster context and call `blog-write`\n\nConstruct the cluster context block (full schema in `references\u002Fexecution-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.\n\n**Evidence provenance propagation.** The cluster context includes this directive\nfor every spoke and the pillar: \"Keep material claims traceable to sources that\nsupport them. Record dates, publisher\u002Ftitle details, retrieval notes,\nmethodology, and limitations when they help identify or interpret the source.\nUse the publication's citation style. Drop unverifiable statistics and replace\ncontradicted ones.\"\n\nThis cascade preserves evidence discipline across batch execution without\nturning a fixed source-record format into a score or gate. See\n`skills\u002Fblog\u002Freferences\u002Fflow-alignment.md`.\n\nThe context also instructs `blog-write` to run autonomously: skip topic clarification, skip outline approval, do not auto-detect template, do not pause.\n\nOutput 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.\n\n### Step 4. Per-post optional hero image\n\nIf `blog-image` is available, call `\u002Fblog image generate` via the Task tool to produce a 16:9 hero image for the post and place it in `cluster-\u003Cslug>\u002Fimages\u002F\u003Cpost-slug>-hero.png`. Delegate provider selection to `blog-image`, prefer current Gemini image models when available, and record the model ID in the scorecard. If image generation is unavailable or fails, log a warning and continue without images. Image generation is non-blocking.\n\n### Step 5. Backward link injection\n\nAfter each post is written:\n\n1. Scan all previously written posts in the cluster directory for `[INTERNAL-LINK: keyword -> filename.md]` markers that reference the just-written post.\n2. Replace each match with a real markdown link: `[keyword](filename.md)`.\n3. Add a cluster metadata block to the post's frontmatter on first pass (`cluster:`, `cluster_role:`, `cluster_group:`).\n\n### Step 6. Failure handling\n\nIf `blog-write` returns a quality-gate failure for any post, stop the batch immediately. Save progress, mark the failed post and all remaining posts as skipped, and tell the user to inspect or retry the failed post manually before continuing. Do not keep generating 5 to 15 posts after a quality failure, because that can create scaled-content-abuse risk.\n\nIf `blog-write` fails before content is generated because of a timeout or runtime error, log the failure in the scorecard and stop unless the user explicitly resumes after checking the cause.\n\nIf the user cancels mid-execution, save progress and note completed posts. On the next `\u002Fblog cluster execute`, detect already-written files and resume from the next unwritten post.\n\n### Step 7. Generate `cluster-scorecard.md`\n\nAfter all attempted posts complete, produce a markdown scorecard covering:\n\n- Per-post status (written, failed, skipped) with file path and word count.\n- Per-post quality score (call `\u002Fblog analyze` on each in parallel via Task) and the cluster average.\n- Cluster cohesion score: a 0 to 100 composite of link reciprocity, intent diversity, template diversity, and keyword coverage (formula in `references\u002Fexecution-workflow.md`).\n- Internal-link audit: outgoing and incoming counts per post, orphan flags, unresolved `[INTERNAL-LINK]` markers.\n- Cannibalization check: any two posts sharing primary keyword, or any pair with greater than 70% keyword overlap. Recommend running `\u002Fblog cannibalization` for a deeper pass.\n- Image generation summary: hero images generated vs. skipped.\n- Recommended next actions: schema generation (`\u002Fblog schema`), per-post SEO validation (`\u002Fblog seo-check`), repurposing (`\u002Fblog repurpose`).\n\n### Step 8. Final report\n\nReturn a concise summary to the user with totals, the scorecard path, and the next-action commands.\n\n---\n\n## Output Artifacts (summary)\n\n| File | Phase | Format |\n|------|-------|--------|\n| `cluster-plan.json` | Plan | JSON |\n| `cluster-map.html` | Plan | Static HTML + inline SVG, no JavaScript |\n| `pillar-\u003Cslug>.md` | Execute | Markdown (or platform-detected format) |\n| `\u003Cspoke-slug>.md` | Execute | Markdown (or platform-detected format) |\n| `images\u002F\u003Cpost-slug>-hero.png` | Execute (optional) | PNG via `blog-image` |\n| `cluster-scorecard.md` | Execute | Markdown |\n\n---\n\n## Quality Gates\n\n| Gate | Check | Action on fail |\n|------|-------|----------------|\n| Cluster minimum | At least 2 clusters with at least 2 posts each | Warn during plan; suggest expansion |\n| Cannibalization | No two posts share primary keyword | Block execution; require plan adjustment |\n| Link completeness | Every post has 2 or more incoming internal links | Warn in scorecard |\n| Length estimate | Pillars often need more depth than spokes | Pass to `blog-write` as optional planning context; never enforce a raw word minimum |\n| Intent diversity | At least 2 distinct intents across clusters | Warn in scorecard |\n| Template diversity | At least 3 distinct templates across the cluster | Warn in scorecard |\n\n---\n\n## Error Handling\n\n| Scenario | Action |\n|----------|--------|\n| Seed keyword too broad (more than 50 keyword variants) | Suggest narrowing the focus before clustering. |\n| Seed keyword too narrow (fewer than 5 keyword variants) | Offer a smaller cluster (pillar plus 2 to 3 spokes) or suggest broadening. |\n| WebSearch unavailable | Produce a draft plan only and require user confirmation, imported keyword data, or SERP data before execution. |\n| `blog-write` fails a quality gate or content generation | Stop the batch, save progress, mark remaining posts as skipped, and require manual inspection or explicit resume. |\n| `blog-write` not installed | Return: \"blog-cluster requires claude-blog. Install it before running this skill.\" |\n| `cluster-plan.json` malformed | Validate JSON and report parse errors with line numbers. |\n| User cancels execution | Save progress; resume on next invocation with already-written posts auto-detected. |\n| Optional image generation unavailable or failed | Skip hero image generation; warn once at start of execute, not per post. |\n\n---\n\n## Differentiation from related claude-blog skills\n\n| Skill | Role | What blog-cluster adds |\n|-------|------|------------------------|\n| `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. |\n| `blog-calendar` | Schedules publication dates around topic clusters | Does not build clusters or write posts; this skill does both. |\n| `blog-cannibalization` | Detects keyword overlap in existing content | Diagnostic only. blog-cluster prevents cannibalization at the planning stage. |\n| `blog-write` | Writes one post at a time | blog-cluster orchestrates many `blog-write` calls with shared cluster context and bidirectional linking. |\n| `blog-outline` | Generates a single SERP-informed outline | blog-cluster generates an outline-equivalent across an entire cluster, then writes the posts. |\n\nblog-cluster is the general contractor: it analyzes the topic, draws the data-driven plan, and builds the entire structure from a single seed keyword.\n",{"name":5,"license":15,"metadata":24,"description":28,"argument-hint":29,"compatibility":30,"user-invokable":31},{"author":25,"version":26,"category":27},"AgriciDaniel","2.2.0","blog","Semantic topic cluster planning and automated execution engine for claude-blog. Performs SERP-based keyword research, groups keywords by search intent and SERP overlap, builds a hub-and-spoke cluster architecture, generates an interactive SVG cluster map, and executes the full cluster by orchestrating blog-write calls with shared cluster context and automatic internal-link injection. Fills the strategy-to-execution gap: blog-strategy plans the blueprint, blog-cluster builds the house. Use when user says \"blog cluster\", \"topic cluster\", \"content cluster\", \"cluster plan\", \"cluster execute\", \"pillar content\", \"hub and spoke\", \"content ecosystem\", \"cluster map\".\n","[plan|execute] [seed-keyword|cluster-plan.json]","Requires Claude Code and claude-blog (provides blog-write, blog-chart, blog-image)",true,"main","af388a77c2fe5e0b4fd879cca4300a0ffa9a6579","skills\u002Fams-imp-blog-cluster",1,"https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis","https:\u002F\u002Fgithub.com\u002FTanNhatCMS","https:\u002F\u002Fmrkiet.name.vn","2026-11-07 14:23:08",{"source":8,"skillName":5,"commands":41,"prompt":51,"zipAvailable":31},[42,45,48],{"tool":43,"command":44},"npx","npx skills add TanNhatCMS\u002Fjarvis --skill blog-cluster",{"tool":46,"command":47},"bunx","bunx skills add TanNhatCMS\u002Fjarvis --skill blog-cluster",{"tool":49,"command":50},"pnpm","pnpm dlx skills add TanNhatCMS\u002Fjarvis --skill blog-cluster","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis\u002Fblob\u002Fmain\u002Fskills\u002Fams-imp-blog-cluster\u002FSKILL.md",[53,59,63],{"path":54,"resourceType":55,"resourceTypeDesc":56,"sizeBytes":57,"sourceHtmlUrl":58},"references\u002Fcluster-architecture.md","REFERENCE","参考文档",7884,"https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis\u002Fblob\u002Fmain\u002Fskills\u002Fams-imp-blog-cluster\u002Freferences\u002Fcluster-architecture.md",{"path":60,"resourceType":55,"resourceTypeDesc":56,"sizeBytes":61,"sourceHtmlUrl":62},"references\u002Fexecution-workflow.md",11092,"https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis\u002Fblob\u002Fmain\u002Fskills\u002Fams-imp-blog-cluster\u002Freferences\u002Fexecution-workflow.md",{"path":64,"resourceType":55,"resourceTypeDesc":56,"sizeBytes":65,"sourceHtmlUrl":66},"references\u002Fsemantic-clustering.md",8325,"https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis\u002Fblob\u002Fmain\u002Fskills\u002Fams-imp-blog-cluster\u002Freferences\u002Fsemantic-clustering.md",[68,75,83,90,97,104],{"skillNo":69,"name":70,"nameZh":6,"description":71,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":72,"categories":73,"keywords":74,"publishedAt":19},"01M7EWYJR0J8DARY7E59Z6Y9MV","web-scrape","Turn web pages into LLM-ready markdown via the Firecrawl API (renders JS, anti-bot handling). Use when browser_open\u002Ffetch fails on JS-heavy pages, or when you need clean markdown\u002Fstructured extraction from a site. Requires FIRECRAWL_API_KEY.","https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis\u002Fblob\u002Fmain\u002Fskills\u002Fweb-scrape\u002FSKILL.md",[],[],{"skillNo":76,"name":77,"nameZh":6,"description":78,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":79,"sourceHtmlUrl":80,"categories":81,"keywords":82,"publishedAt":19},"01M7EWYBN5P6F2AJG3FR24T9BP","design-taste-frontend","tasteskill v2, the anti-slop frontend skill. Art direction for websites, distilled from a study of 247 curated gallery sites (One Page Love, Awwwards Sites of the Day). Deals researched directions (idea, composition, type, palette, photographs, section blueprints) from a library, enforces a strict no-slop house style with a real-browser inspector, and runs a look-and-fix loop on screenshots. Use this whenever you build, design, redesign, restyle, polish or review a landing page, marketing site, portfolio, product page, one-pager, hero or any public-facing web page, in any stack (HTML\u002FCSS, React, Next.js, Vue, Svelte, Astro, Tailwind) — also when the user says a page looks generic, templated, \"AI-made\", sloppy, or asks to make it look designed, premium, award-level or less like every other site. Not for dashboards or app screens.","https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis\u002Fblob\u002Fmain\u002Fskills\u002Ftaste-design\u002FLICENSE","https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis\u002Fblob\u002Fmain\u002Fskills\u002Ftaste-design\u002FSKILL.md",[],[],{"skillNo":84,"name":85,"nameZh":6,"description":86,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":87,"categories":88,"keywords":89,"publishedAt":19},"01M7EWY8FQFNQ2GB7NZZWGSSDX","test-driven-development","Use when implementing any feature or bugfix, before writing implementation code","https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis\u002Fblob\u002Fmain\u002Fskills\u002Fsuperpowers-tdd\u002FSKILL.md",[],[],{"skillNo":91,"name":92,"nameZh":6,"description":93,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":94,"categories":95,"keywords":96,"publishedAt":19},"01M7EWY5SHBFRH1HDHVWM9K7TV","systematic-debugging","Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes","https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis\u002Fblob\u002Fmain\u002Fskills\u002Fsuperpowers-debugging\u002FSKILL.md",[],[],{"skillNo":98,"name":99,"nameZh":6,"description":100,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":101,"categories":102,"keywords":103,"publishedAt":19},"01M7EWY3NM0Z0JAKS7HDZ24K66","requesting-code-review","Use when completing tasks, implementing major features, or before merging to verify work meets requirements","https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis\u002Fblob\u002Fmain\u002Fskills\u002Fsuperpowers-code-review\u002FSKILL.md",[],[],{"skillNo":105,"name":106,"nameZh":6,"description":107,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":108,"categories":109,"keywords":110,"publishedAt":19},"01M7EWY0WRMWX4M9ZFY0WKC593","brainstorming","You MUST use this before any creative work - creating features, building components, adding functionality, modifying behavior, or planning anything new, in software or out of it (a talk, a business, a renovation).","https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis\u002Fblob\u002Fmain\u002Fskills\u002Fsuperpowers-brainstorming\u002FSKILL.md",[],[]]