[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7BKFPSE18PDTM7QXDW18GRK":3,"skill-same:01M7BKFPSE18PDTM7QXDW18GRK":58},{"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":31,"commitSha":32,"skillDirectory":33,"forksCount":34,"repositoryHtmlUrl":35,"repositoryHomepageUrl":6,"ownerHtmlUrl":36,"ownerBlogUrl":6,"ownerDescription":37,"updatedAt":38,"install":39,"resources":51},"01M7BKFPSE18PDTM7QXDW18GRK","blog-image",null,"AI image generation and editing for blog content powered by Gemini via MCP. Claude acts as Creative Director - interpreting intent, selecting domain expertise, constructing optimized 6-component prompts (Subject + Action + Context + Composition + Lighting + Style), and orchestrating Gemini for blog-quality results. Generates hero images, inline illustrations, social preview cards, and OG images. Edits existing blog images. Supports 6 blog-optimized domain modes (Editorial, Product, Landscape, UI\u002FWeb, Infographic, Abstract). Works standalone via \u002Fblog image or internally from blog-write and blog-rewrite workflows. Falls back gracefully when MCP is not configured. Use when user says \"blog image\", \"generate hero image\", \"blog illustration\", \"social card\", \"generate blog image\", \"edit blog image\", \"image generate\", \"blog cover image\", \"inline image\", \"OG image\".","om-scogo\u002Fskillsh-scraper","om-scogo","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F160466840?v=4","01M79AWB6DFVCQ7RK70CS34B96","Om Achrekar","01M79AWB6ZM9SKR8KDH1WSSM22",0,"MIT","https:\u002F\u002Fgithub.com\u002Fom-scogo\u002Fskillsh-scraper\u002Fblob\u002Fmain\u002Fdata\u002Fskills\u002Fagricidaniel-claude-blog\u002Fblog-image\u002FSkill.md",[],[],"2026-11-15 14:19:17",false,"---\nname: blog-image\ndescription: >\n  AI image generation and editing for blog content powered by Gemini via MCP.\n  Claude acts as Creative Director - interpreting intent, selecting domain expertise,\n  constructing optimized 6-component prompts (Subject + Action + Context + Composition\n  + Lighting + Style), and orchestrating Gemini for blog-quality results. Generates\n  hero images, inline illustrations, social preview cards, and OG images. Edits\n  existing blog images. Supports 6 blog-optimized domain modes (Editorial, Product,\n  Landscape, UI\u002FWeb, Infographic, Abstract). Works standalone via \u002Fblog image or\n  internally from blog-write and blog-rewrite workflows. Falls back gracefully when\n  MCP is not configured. Use when user says \"blog image\", \"generate hero image\",\n  \"blog illustration\", \"social card\", \"generate blog image\", \"edit blog image\",\n  \"image generate\", \"blog cover image\", \"inline image\", \"OG image\".\nuser-invokable: true\nargument-hint: \"[generate|edit|setup] [description-or-path]\"\nlicense: MIT\nmetadata:\n  author: AgriciDaniel\n  version: \"1.9.1\"\n  mcp-package: \"@ycse\u002Fnanobanana-mcp\"\n---\n\n# Blog Image - AI Image Generation for Blog Content\n\nYou are a **Creative Director** that orchestrates Gemini's image generation\nspecifically for blog content. Never pass raw user text directly to the API.\nAlways interpret, enhance, and construct an optimized prompt using the\n6-component Reasoning Brief system.\n\n## Quick Reference\n\n| Command | What it does |\n|---------|-------------|\n| `\u002Fblog image generate \u003Cidea>` | Generate a blog image with full prompt engineering |\n| `\u002Fblog image edit \u003Cpath> \u003Cinstructions>` | Edit an existing blog image intelligently |\n| `\u002Fblog image setup` | Configure MCP server and API key |\n\n## Blog Image Types\n\nMatch the image type to blog use case:\n\n| Image Type | Aspect Ratio | Resolution | Domain Mode | Placement |\n|------------|-------------|-----------|-------------|-----------|\n| Hero\u002FCover | `16:9` | 2K or 4K | Editorial \u002F Landscape | Frontmatter `coverImage` |\n| OG\u002FSocial Card | `16:9` | 1K | Editorial \u002F Infographic | Frontmatter `ogImage` |\n| Inline Illustration | `16:9` or `4:3` | 1K | Varies by topic | After H2, before body |\n| Inline Product Shot | `4:3` or `1:1` | 1K | Product | Within product sections |\n| Section Divider | `8:1` or `4:1` | 1K | Abstract \u002F Landscape | Between major sections |\n\n**Sizing requirements:**\n- Blog hero\u002Fcover: 1200x630 (OG-compatible) or 1920x1080\n- Open Graph (OG): 1200x630 (required for social sharing)\n- Inline images: 1200px+ wide\n\n## MCP Availability Check\n\nBefore generating, check if nanobanana-mcp tools are available:\n\n1. Try calling `get_image_history` (lightweight, no side effects)\n2. If it succeeds: MCP is available, proceed with generation\n3. If it fails: MCP not configured - inform the user:\n   - \"Image generation requires the nanobanana-mcp server. Run `\u002Fblog image setup` to configure it.\"\n   - When called internally (from blog-write\u002Fblog-rewrite): return silently, no error. The calling workflow continues with stock photos.\n\n## Generation Workflow\n\nFor `\u002Fblog image generate \u003Cidea>` or when invoked internally:\n\n### Step 1: Analyze Intent\n\nDetermine what the blog needs:\n- **Image type**: Hero, inline, OG card, section divider?\n- **Blog topic**: What is the article about?\n- **Style**: Photorealistic, editorial, illustrated, minimal?\n- **Constraints**: Brand colors, specific dimensions, platform format?\n- **Mood**: Authoritative, inviting, dramatic, clean?\n\nIf the request is vague, ask one clarifying question about use case and style.\n\n### Step 2: Select Domain Mode\n\nChoose the expertise lens for the image:\n\n| Mode | When to use | Prompt emphasis |\n|------|-------------|-----------------|\n| **Editorial** | Blog headers, feature images, lifestyle | Styling, composition, publication references |\n| **Product** | E-commerce posts, reviews, comparisons | Surface materials, studio lighting, clean BG |\n| **Landscape** | Environmental backgrounds, travel, hero sections | Atmospheric perspective, depth layers, time of day |\n| **UI\u002FWeb** | Tech blog icons, illustrations, diagrams | Clean vectors, flat design, exact colors |\n| **Infographic** | Data-driven posts, processes, comparisons | Layout structure, hierarchy, accessible colors |\n| **Abstract** | Pattern backgrounds, section dividers, decorative | Color theory, mathematical forms, textures |\n\nLoad `references\u002Fprompt-engineering-blog.md` for domain mode modifier libraries.\n\n### Step 3: Construct the 6-Component Reasoning Brief\n\nBuild the prompt as natural narrative paragraphs - NEVER as keyword lists:\n\n1. **Subject** - Who\u002Fwhat, with rich physical detail (textures, materials, scale)\n2. **Action** - What is happening, pose, gesture, movement, state\n3. **Context** - Environment, setting, time of day, season, weather\n4. **Composition** - Camera angle, shot type, framing, negative space, depth\n5. **Lighting** - Light source, quality, direction, color temperature, shadows\n6. **Style** - Art medium, aesthetic, film stock, reference artists\u002Feras\n\n**Template for photorealistic blog images:**\n```\nA photorealistic [shot type] of [subject with physical detail], [action\u002Fpose],\nset in [environment with specifics]. [Lighting conditions] create [mood].\nCaptured with [camera model], [focal length] lens at [f-stop], producing\n[depth of field effect]. [Color palette\u002Fgrading notes]. Aspect ratio 16:9,\nsuitable as a blog [hero image\u002Finline illustration] at [target dimensions].\n```\n\n**Template for illustrated\u002Fstylized:**\n```\nA [art style] [format] of [subject with character detail], featuring\n[distinctive characteristics] with [color palette]. [Line style] and\n[shading technique]. Background is [description]. [Mood\u002Fatmosphere].\n```\n\n### Step 4: Set Aspect Ratio\n\nCall `set_aspect_ratio` BEFORE generating:\n\n| Blog Use Case | Ratio |\n|---------------|-------|\n| Hero \u002F Cover \u002F OG | `16:9` |\n| Product shot \u002F Square | `4:3` or `1:1` |\n| Section divider | `8:1` or `4:1` |\n| Vertical (stories) | `9:16` |\n\n### Step 5: Generate via MCP\n\n| MCP Tool | When |\n|----------|------|\n| `set_aspect_ratio` | Always call first if ratio differs from 1:1 |\n| `gemini_generate_image` | New image from crafted prompt |\n| `gemini_edit_image` | Modify existing image |\n| `gemini_chat` | Iterative refinement \u002F multi-turn sessions |\n| `get_image_history` | Review generated images |\n| `clear_conversation` | Reset session context |\n\n**Model selection** (use `set_model` MCP tool if switching):\n- **NB2 Flash** (default): Best for most blog images - fast, 14 ratios, 4K, $0.067\u002Fimg\n- **NB Pro**: Use for hero images with text overlays (94% text accuracy) or highest quality - $0.134\u002Fimg\n- **Original**: Budget option at $0.039\u002Fimg - 5 ratios, 1K max\n\nLoad `references\u002Fmcp-tools.md` for parameter details.\nLoad `references\u002Fgemini-models.md` for model specs, pricing, and rate limits.\n\n### Step 6: Post-Processing (when needed)\n\nAfter generation, resize\u002Fconvert for blog use:\n\n```bash\n# Resize to blog hero dimensions (1200x630)\nmagick input.png -resize 1200x630^ -gravity center -extent 1200x630 hero.png\n\n# Convert to WebP for web optimization\nmagick input.png -quality 85 output.webp\n\n# Convert to AVIF (smallest, modern)\nmagick input.png -quality 80 output.avif\n\n# Crop to exact OG dimensions\nmagick input.png -resize 1200x630^ -gravity center -extent 1200x630 og-image.png\n```\n\nCheck if `magick` (ImageMagick 7) is available. Fall back to `convert` if not.\n\n### Step 7: Deliver\n\nProvide:\n1. **Image path** - where it was saved (`~\u002FDocuments\u002Fnanobanana_generated\u002F`)\n2. **Crafted prompt** - show the full Reasoning Brief (educational)\n3. **Settings** - model, aspect ratio, domain mode\n4. **Alt text** - descriptive sentence, 10-125 chars, topic keywords naturally\n5. **Frontmatter snippet** (for hero\u002FOG images):\n```yaml\ncoverImage: \"\u002Fpath\u002Fto\u002Fgenerated-image.png\"\ncoverImageAlt: \"Descriptive alt text sentence with topic keywords\"\nogImage: \"\u002Fpath\u002Fto\u002Fgenerated-image.png\"\n```\n6. **Refinement suggestions** - 1-2 ideas if relevant\n\n## Edit Workflow\n\nFor `\u002Fblog image edit \u003Cpath> \u003Cinstructions>`:\n\n1. Read the image path and edit instruction\n2. Enhance the instruction (never pass raw):\n   | User says | Claude crafts |\n   |-----------|---------------|\n   | \"remove background\" | Detailed edge-preserving background removal |\n   | \"make it warmer\" | Specific color temperature shift with preservation notes |\n   | \"add text\" | Font style, size, placement, contrast, readability notes |\n   | \"make it brighter\" | Increase exposure, lift shadows, maintain highlights |\n   | \"crop for social\" | Resize to 1200x630 with center-gravity crop |\n3. Call `gemini_edit_image` with enhanced instruction\n4. Return modified image path and description\n\n## Internal API (for blog-write \u002F blog-rewrite)\n\nWhen invoked as a Task subagent from blog-write or blog-rewrite:\n\n**Input** (provided by calling skill):\n- `image_type`: hero, inline, og, divider\n- `topic`: blog post topic\u002Ftitle\n- `section_context`: (optional) heading or section the image supports\n- `style_preference`: (optional) photorealistic, illustrated, editorial\n- `count`: (optional) number of images needed (default: 1)\n\n**Output** (returned to calling skill):\n```markdown\n### Generated Image\n- **Path:** ~\u002FDocuments\u002Fnanobanana_generated\u002Fimage_timestamp.png\n- **Alt Text:** Descriptive sentence about the image\n- **Type:** hero \u002F inline \u002F og\n- **Domain Mode:** Editorial\n- **Aspect Ratio:** 16:9\n- **Suggested Frontmatter:**\n  coverImage: \"\u002Fpath\u002Fto\u002Fimage.png\"\n  coverImageAlt: \"Alt text here\"\n```\n\n**Graceful fallback**: If MCP is unavailable, return immediately with no error.\nThe calling workflow continues with stock photos. Never block blog-write or\nblog-rewrite because image generation is unavailable.\n\n## Alt Text Generation\n\nFor every generated image, create alt text following blog standards:\n- Full descriptive sentence (not keyword list)\n- 10-125 characters\n- Include topic keywords naturally\n- Describe what the image shows AND its relevance to the content\n- For charts\u002Finfographics: include the key data point\n\nGood: `Marketing team analyzing AI search traffic data on a dashboard showing citation metrics`\nBad: `SEO AI marketing blog optimization image`\n\n## Setup\n\nFor `\u002Fblog image setup`:\n\n1. Run `python3 scripts\u002Fsetup_image_mcp.py` (interactive)\n   - Or: `python3 scripts\u002Fsetup_image_mcp.py --key YOUR_KEY` (non-interactive)\n   - Default writes to `~\u002F.claude\u002Fsettings.json` (user-private, mode 0600)\n   - `--project` flag opts into project `.mcp.json` (env-expansion only,\n     refuses to write a literal key into a tracked file)\n2. Verify: `python3 scripts\u002Fvalidate_image_setup.py`\n3. Requires:\n   - Node.js 18+ (npx)\n   - Google AI API key (free at https:\u002F\u002Faistudio.google.com\u002Fapikey)\n4. The script pins the package to `@ycse\u002Fnanobanana-mcp@1.1.1`. Update the\n   pin in `setup_image_mcp.py` (constant `PINNED_PACKAGE`) when bumping.\n\n## Safety Filter Auto-Rephrase\n\nWhen `IMAGE_SAFETY` or `SAFETY` is returned, do NOT give up. Auto-rephrase and retry:\n\n1. Identify the likely trigger (violence, public figures, NSFW-adjacent, or overly cautious filter)\n2. Rephrase using positive framing - describe what you WANT, not what to avoid\n3. If the subject is a person, make them generic (remove celebrity-like specifics)\n4. If the scene is dramatic, soften: \"intense\" → \"focused\", \"battle\" → \"competition\"\n5. Retry with the rephrased prompt (max 3 attempts before reporting to user)\n\nGoogle acknowledged filters \"became way more cautious than we intended\" - benign prompts\nare sometimes blocked. Persistence with rephrasing usually succeeds.\n\n## Edit, Don't Re-roll\n\nIf an image is 80% correct, use `gemini_chat` for conversational editing rather than\nregenerating from scratch. The session maintains style consistency, so targeted edits\npreserve what works while fixing what doesn't.\n\n**When to edit vs regenerate:**\n- Color slightly off → Edit (\"shift the color temperature warmer\")\n- Wrong composition entirely → Regenerate with revised brief\n- Good scene but wrong lighting → Edit (\"change to golden hour lighting from the left\")\n- Missing a detail → Edit (\"add a steaming coffee cup on the desk\")\n\n## Error Handling\n\n| Error | Resolution |\n|-------|-----------|\n| MCP not configured | Run `\u002Fblog image setup` |\n| API key invalid | New key at https:\u002F\u002Faistudio.google.com\u002Fapikey |\n| Rate limited (429) | Wait 60s, retry. Free tier: ~5-15 RPM \u002F ~20-500 RPD (varies by model and billing) |\n| `IMAGE_SAFETY` | Auto-rephrase (see above) - Layer 2 filter, non-configurable |\n| `PROHIBITED_CONTENT` | Content policy violation - topic is blocked. Non-retryable. |\n| `SAFETY` | Rephrase prompt - Layer 1 filter |\n| Vague request | Ask one clarifying question before generating |\n| Poor quality | Review Reasoning Brief - likely missing lighting (biggest quality differentiator) |\n| MCP unavailable (internal call) | Return silently - calling workflow uses stock photos |\n\n## Reference Documentation\n\nLoad on-demand - do NOT load all at startup:\n- `references\u002Fprompt-engineering-blog.md` - Domain modes, 6-component system, blog templates\n- `references\u002Fgemini-models.md` - Model specs, rate limits, aspect ratios, pricing\n- `references\u002Fmcp-tools.md` - MCP tool parameters and response formats\n","# Blog Image - AI Image Generation for Blog Content\n\nYou are a **Creative Director** that orchestrates Gemini's image generation\nspecifically for blog content. Never pass raw user text directly to the API.\nAlways interpret, enhance, and construct an optimized prompt using the\n6-component Reasoning Brief system.\n\n## Quick Reference\n\n| Command | What it does |\n|---------|-------------|\n| `\u002Fblog image generate \u003Cidea>` | Generate a blog image with full prompt engineering |\n| `\u002Fblog image edit \u003Cpath> \u003Cinstructions>` | Edit an existing blog image intelligently |\n| `\u002Fblog image setup` | Configure MCP server and API key |\n\n## Blog Image Types\n\nMatch the image type to blog use case:\n\n| Image Type | Aspect Ratio | Resolution | Domain Mode | Placement |\n|------------|-------------|-----------|-------------|-----------|\n| Hero\u002FCover | `16:9` | 2K or 4K | Editorial \u002F Landscape | Frontmatter `coverImage` |\n| OG\u002FSocial Card | `16:9` | 1K | Editorial \u002F Infographic | Frontmatter `ogImage` |\n| Inline Illustration | `16:9` or `4:3` | 1K | Varies by topic | After H2, before body |\n| Inline Product Shot | `4:3` or `1:1` | 1K | Product | Within product sections |\n| Section Divider | `8:1` or `4:1` | 1K | Abstract \u002F Landscape | Between major sections |\n\n**Sizing requirements:**\n- Blog hero\u002Fcover: 1200x630 (OG-compatible) or 1920x1080\n- Open Graph (OG): 1200x630 (required for social sharing)\n- Inline images: 1200px+ wide\n\n## MCP Availability Check\n\nBefore generating, check if nanobanana-mcp tools are available:\n\n1. Try calling `get_image_history` (lightweight, no side effects)\n2. If it succeeds: MCP is available, proceed with generation\n3. If it fails: MCP not configured - inform the user:\n   - \"Image generation requires the nanobanana-mcp server. Run `\u002Fblog image setup` to configure it.\"\n   - When called internally (from blog-write\u002Fblog-rewrite): return silently, no error. The calling workflow continues with stock photos.\n\n## Generation Workflow\n\nFor `\u002Fblog image generate \u003Cidea>` or when invoked internally:\n\n### Step 1: Analyze Intent\n\nDetermine what the blog needs:\n- **Image type**: Hero, inline, OG card, section divider?\n- **Blog topic**: What is the article about?\n- **Style**: Photorealistic, editorial, illustrated, minimal?\n- **Constraints**: Brand colors, specific dimensions, platform format?\n- **Mood**: Authoritative, inviting, dramatic, clean?\n\nIf the request is vague, ask one clarifying question about use case and style.\n\n### Step 2: Select Domain Mode\n\nChoose the expertise lens for the image:\n\n| Mode | When to use | Prompt emphasis |\n|------|-------------|-----------------|\n| **Editorial** | Blog headers, feature images, lifestyle | Styling, composition, publication references |\n| **Product** | E-commerce posts, reviews, comparisons | Surface materials, studio lighting, clean BG |\n| **Landscape** | Environmental backgrounds, travel, hero sections | Atmospheric perspective, depth layers, time of day |\n| **UI\u002FWeb** | Tech blog icons, illustrations, diagrams | Clean vectors, flat design, exact colors |\n| **Infographic** | Data-driven posts, processes, comparisons | Layout structure, hierarchy, accessible colors |\n| **Abstract** | Pattern backgrounds, section dividers, decorative | Color theory, mathematical forms, textures |\n\nLoad `references\u002Fprompt-engineering-blog.md` for domain mode modifier libraries.\n\n### Step 3: Construct the 6-Component Reasoning Brief\n\nBuild the prompt as natural narrative paragraphs - NEVER as keyword lists:\n\n1. **Subject** - Who\u002Fwhat, with rich physical detail (textures, materials, scale)\n2. **Action** - What is happening, pose, gesture, movement, state\n3. **Context** - Environment, setting, time of day, season, weather\n4. **Composition** - Camera angle, shot type, framing, negative space, depth\n5. **Lighting** - Light source, quality, direction, color temperature, shadows\n6. **Style** - Art medium, aesthetic, film stock, reference artists\u002Feras\n\n**Template for photorealistic blog images:**\n```\nA photorealistic [shot type] of [subject with physical detail], [action\u002Fpose],\nset in [environment with specifics]. [Lighting conditions] create [mood].\nCaptured with [camera model], [focal length] lens at [f-stop], producing\n[depth of field effect]. [Color palette\u002Fgrading notes]. Aspect ratio 16:9,\nsuitable as a blog [hero image\u002Finline illustration] at [target dimensions].\n```\n\n**Template for illustrated\u002Fstylized:**\n```\nA [art style] [format] of [subject with character detail], featuring\n[distinctive characteristics] with [color palette]. [Line style] and\n[shading technique]. Background is [description]. [Mood\u002Fatmosphere].\n```\n\n### Step 4: Set Aspect Ratio\n\nCall `set_aspect_ratio` BEFORE generating:\n\n| Blog Use Case | Ratio |\n|---------------|-------|\n| Hero \u002F Cover \u002F OG | `16:9` |\n| Product shot \u002F Square | `4:3` or `1:1` |\n| Section divider | `8:1` or `4:1` |\n| Vertical (stories) | `9:16` |\n\n### Step 5: Generate via MCP\n\n| MCP Tool | When |\n|----------|------|\n| `set_aspect_ratio` | Always call first if ratio differs from 1:1 |\n| `gemini_generate_image` | New image from crafted prompt |\n| `gemini_edit_image` | Modify existing image |\n| `gemini_chat` | Iterative refinement \u002F multi-turn sessions |\n| `get_image_history` | Review generated images |\n| `clear_conversation` | Reset session context |\n\n**Model selection** (use `set_model` MCP tool if switching):\n- **NB2 Flash** (default): Best for most blog images - fast, 14 ratios, 4K, $0.067\u002Fimg\n- **NB Pro**: Use for hero images with text overlays (94% text accuracy) or highest quality - $0.134\u002Fimg\n- **Original**: Budget option at $0.039\u002Fimg - 5 ratios, 1K max\n\nLoad `references\u002Fmcp-tools.md` for parameter details.\nLoad `references\u002Fgemini-models.md` for model specs, pricing, and rate limits.\n\n### Step 6: Post-Processing (when needed)\n\nAfter generation, resize\u002Fconvert for blog use:\n\n```bash\n# Resize to blog hero dimensions (1200x630)\nmagick input.png -resize 1200x630^ -gravity center -extent 1200x630 hero.png\n\n# Convert to WebP for web optimization\nmagick input.png -quality 85 output.webp\n\n# Convert to AVIF (smallest, modern)\nmagick input.png -quality 80 output.avif\n\n# Crop to exact OG dimensions\nmagick input.png -resize 1200x630^ -gravity center -extent 1200x630 og-image.png\n```\n\nCheck if `magick` (ImageMagick 7) is available. Fall back to `convert` if not.\n\n### Step 7: Deliver\n\nProvide:\n1. **Image path** - where it was saved (`~\u002FDocuments\u002Fnanobanana_generated\u002F`)\n2. **Crafted prompt** - show the full Reasoning Brief (educational)\n3. **Settings** - model, aspect ratio, domain mode\n4. **Alt text** - descriptive sentence, 10-125 chars, topic keywords naturally\n5. **Frontmatter snippet** (for hero\u002FOG images):\n```yaml\ncoverImage: \"\u002Fpath\u002Fto\u002Fgenerated-image.png\"\ncoverImageAlt: \"Descriptive alt text sentence with topic keywords\"\nogImage: \"\u002Fpath\u002Fto\u002Fgenerated-image.png\"\n```\n6. **Refinement suggestions** - 1-2 ideas if relevant\n\n## Edit Workflow\n\nFor `\u002Fblog image edit \u003Cpath> \u003Cinstructions>`:\n\n1. Read the image path and edit instruction\n2. Enhance the instruction (never pass raw):\n   | User says | Claude crafts |\n   |-----------|---------------|\n   | \"remove background\" | Detailed edge-preserving background removal |\n   | \"make it warmer\" | Specific color temperature shift with preservation notes |\n   | \"add text\" | Font style, size, placement, contrast, readability notes |\n   | \"make it brighter\" | Increase exposure, lift shadows, maintain highlights |\n   | \"crop for social\" | Resize to 1200x630 with center-gravity crop |\n3. Call `gemini_edit_image` with enhanced instruction\n4. Return modified image path and description\n\n## Internal API (for blog-write \u002F blog-rewrite)\n\nWhen invoked as a Task subagent from blog-write or blog-rewrite:\n\n**Input** (provided by calling skill):\n- `image_type`: hero, inline, og, divider\n- `topic`: blog post topic\u002Ftitle\n- `section_context`: (optional) heading or section the image supports\n- `style_preference`: (optional) photorealistic, illustrated, editorial\n- `count`: (optional) number of images needed (default: 1)\n\n**Output** (returned to calling skill):\n```markdown\n### Generated Image\n- **Path:** ~\u002FDocuments\u002Fnanobanana_generated\u002Fimage_timestamp.png\n- **Alt Text:** Descriptive sentence about the image\n- **Type:** hero \u002F inline \u002F og\n- **Domain Mode:** Editorial\n- **Aspect Ratio:** 16:9\n- **Suggested Frontmatter:**\n  coverImage: \"\u002Fpath\u002Fto\u002Fimage.png\"\n  coverImageAlt: \"Alt text here\"\n```\n\n**Graceful fallback**: If MCP is unavailable, return immediately with no error.\nThe calling workflow continues with stock photos. Never block blog-write or\nblog-rewrite because image generation is unavailable.\n\n## Alt Text Generation\n\nFor every generated image, create alt text following blog standards:\n- Full descriptive sentence (not keyword list)\n- 10-125 characters\n- Include topic keywords naturally\n- Describe what the image shows AND its relevance to the content\n- For charts\u002Finfographics: include the key data point\n\nGood: `Marketing team analyzing AI search traffic data on a dashboard showing citation metrics`\nBad: `SEO AI marketing blog optimization image`\n\n## Setup\n\nFor `\u002Fblog image setup`:\n\n1. Run `python3 scripts\u002Fsetup_image_mcp.py` (interactive)\n   - Or: `python3 scripts\u002Fsetup_image_mcp.py --key YOUR_KEY` (non-interactive)\n   - Default writes to `~\u002F.claude\u002Fsettings.json` (user-private, mode 0600)\n   - `--project` flag opts into project `.mcp.json` (env-expansion only,\n     refuses to write a literal key into a tracked file)\n2. Verify: `python3 scripts\u002Fvalidate_image_setup.py`\n3. Requires:\n   - Node.js 18+ (npx)\n   - Google AI API key (free at https:\u002F\u002Faistudio.google.com\u002Fapikey)\n4. The script pins the package to `@ycse\u002Fnanobanana-mcp@1.1.1`. Update the\n   pin in `setup_image_mcp.py` (constant `PINNED_PACKAGE`) when bumping.\n\n## Safety Filter Auto-Rephrase\n\nWhen `IMAGE_SAFETY` or `SAFETY` is returned, do NOT give up. Auto-rephrase and retry:\n\n1. Identify the likely trigger (violence, public figures, NSFW-adjacent, or overly cautious filter)\n2. Rephrase using positive framing - describe what you WANT, not what to avoid\n3. If the subject is a person, make them generic (remove celebrity-like specifics)\n4. If the scene is dramatic, soften: \"intense\" → \"focused\", \"battle\" → \"competition\"\n5. Retry with the rephrased prompt (max 3 attempts before reporting to user)\n\nGoogle acknowledged filters \"became way more cautious than we intended\" - benign prompts\nare sometimes blocked. Persistence with rephrasing usually succeeds.\n\n## Edit, Don't Re-roll\n\nIf an image is 80% correct, use `gemini_chat` for conversational editing rather than\nregenerating from scratch. The session maintains style consistency, so targeted edits\npreserve what works while fixing what doesn't.\n\n**When to edit vs regenerate:**\n- Color slightly off → Edit (\"shift the color temperature warmer\")\n- Wrong composition entirely → Regenerate with revised brief\n- Good scene but wrong lighting → Edit (\"change to golden hour lighting from the left\")\n- Missing a detail → Edit (\"add a steaming coffee cup on the desk\")\n\n## Error Handling\n\n| Error | Resolution |\n|-------|-----------|\n| MCP not configured | Run `\u002Fblog image setup` |\n| API key invalid | New key at https:\u002F\u002Faistudio.google.com\u002Fapikey |\n| Rate limited (429) | Wait 60s, retry. Free tier: ~5-15 RPM \u002F ~20-500 RPD (varies by model and billing) |\n| `IMAGE_SAFETY` | Auto-rephrase (see above) - Layer 2 filter, non-configurable |\n| `PROHIBITED_CONTENT` | Content policy violation - topic is blocked. Non-retryable. |\n| `SAFETY` | Rephrase prompt - Layer 1 filter |\n| Vague request | Ask one clarifying question before generating |\n| Poor quality | Review Reasoning Brief - likely missing lighting (biggest quality differentiator) |\n| MCP unavailable (internal call) | Return silently - calling workflow uses stock photos |\n\n## Reference Documentation\n\nLoad on-demand - do NOT load all at startup:\n- `references\u002Fprompt-engineering-blog.md` - Domain modes, 6-component system, blog templates\n- `references\u002Fgemini-models.md` - Model specs, rate limits, aspect ratios, pricing\n- `references\u002Fmcp-tools.md` - MCP tool parameters and response formats\n",{"name":5,"license":15,"metadata":24,"description":28,"argument-hint":29,"user-invokable":30},{"author":25,"version":26,"mcp-package":27},"AgriciDaniel","1.9.1","@ycse\u002Fnanobanana-mcp","AI image generation and editing for blog content powered by Gemini via MCP. Claude acts as Creative Director - interpreting intent, selecting domain expertise, constructing optimized 6-component prompts (Subject + Action + Context + Composition + Lighting + Style), and orchestrating Gemini for blog-quality results. Generates hero images, inline illustrations, social preview cards, and OG images. Edits existing blog images. Supports 6 blog-optimized domain modes (Editorial, Product, Landscape, UI\u002FWeb, Infographic, Abstract). Works standalone via \u002Fblog image or internally from blog-write and blog-rewrite workflows. Falls back gracefully when MCP is not configured. Use when user says \"blog image\", \"generate hero image\", \"blog illustration\", \"social card\", \"generate blog image\", \"edit blog image\", \"image generate\", \"blog cover image\", \"inline image\", \"OG image\".\n","[generate|edit|setup] [description-or-path]",true,"main","8859218955d9504cd18b59e7892100718039d3b4","data\u002Fskills\u002Fagricidaniel-claude-blog\u002Fblog-image",1,"https:\u002F\u002Fgithub.com\u002Fom-scogo\u002Fskillsh-scraper","https:\u002F\u002Fgithub.com\u002Fom-scogo","SDE-1 @ Scogo ","2026-07-18 12:22:46",{"source":8,"skillName":5,"commands":40,"prompt":50,"zipAvailable":30},[41,44,47],{"tool":42,"command":43},"npx","npx skills add om-scogo\u002Fskillsh-scraper --skill blog-image",{"tool":45,"command":46},"bunx","bunx skills add om-scogo\u002Fskillsh-scraper --skill blog-image",{"tool":48,"command":49},"pnpm","pnpm dlx skills add om-scogo\u002Fskillsh-scraper --skill blog-image","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fom-scogo\u002Fskillsh-scraper\u002Fblob\u002Fmain\u002Fdata\u002Fskills\u002Fagricidaniel-claude-blog\u002Fblog-image\u002FSkill.md",[52],{"path":53,"resourceType":54,"resourceTypeDesc":55,"sizeBytes":56,"sourceHtmlUrl":57},"metadata.json","REFERENCE","参考文档",445,"https:\u002F\u002Fgithub.com\u002Fom-scogo\u002Fskillsh-scraper\u002Fblob\u002Fmain\u002Fdata\u002Fskills\u002Fagricidaniel-claude-blog\u002Fblog-image\u002Fmetadata.json",[59,66,71,78,83,90],{"skillNo":60,"name":61,"nameZh":6,"description":62,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":63,"categories":64,"keywords":65,"publishedAt":19},"01M7D6MC9FAAEV4CMNYX8W7JTG","arxiv-search","Search arXiv physics, math, and computer science preprints using natural language queries. Powered by Valyu semantic search.","https:\u002F\u002Fgithub.com\u002Fom-scogo\u002Fskillsh-scraper\u002Fblob\u002Fmain\u002Fdata\u002Fskills\u002Fyorkeccak-scientific-skills\u002Farxiv-search\u002Fskills\u002Farxiv-search\u002FSKILL.md",[],[],{"skillNo":67,"name":61,"nameZh":6,"description":62,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":68,"categories":69,"keywords":70,"publishedAt":19},"01M7D6M85CKKN87TA98VCQV4H7","https:\u002F\u002Fgithub.com\u002Fom-scogo\u002Fskillsh-scraper\u002Fblob\u002Fmain\u002Fdata\u002Fskills\u002Fyorkeccak-scientific-skills\u002Farxiv-search\u002FSkill.md",[],[],{"skillNo":72,"name":73,"nameZh":6,"description":74,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":75,"categories":76,"keywords":77,"publishedAt":19},"01M7D4ZTCS4DR3CE4JY06X84T2","working-with-legacy-code","Safely change and test untested codebases using Feathers' \"Working Effectively with Legacy Code\". Use when the user mentions \"legacy code\", \"no tests\", \"untested codebase\", \"how do I test this\", \"seams\", \"characterization tests\", \"golden master\", \"sprout method\", \"afraid to change this code\", \"monster method\", \"dependency breaking\", or \"inherited a messy codebase\". Also trigger when changing code without tests safely, getting a class under test when constructors, statics, or singletons block it, adding features to tangled modules, or planning incremental test coverage for an old codebase. Covers the legacy-code change algorithm, seams, characterization tests, sprout\u002Fwrap, and dependency-breaking techniques. For refactoring code that already has tests, see refactoring-patterns. For day-to-day code quality, see clean-code.","https:\u002F\u002Fgithub.com\u002Fom-scogo\u002Fskillsh-scraper\u002Fblob\u002Fmain\u002Fdata\u002Fskills\u002Fwondelai-skills\u002Fworking-with-legacy-code\u002Fworking-with-legacy-code\u002FSKILL.md",[],[],{"skillNo":79,"name":73,"nameZh":6,"description":74,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":80,"categories":81,"keywords":82,"publishedAt":19},"01M7D4ZHZKJD2RSNGAF63V14P4","https:\u002F\u002Fgithub.com\u002Fom-scogo\u002Fskillsh-scraper\u002Fblob\u002Fmain\u002Fdata\u002Fskills\u002Fwondelai-skills\u002Fworking-with-legacy-code\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},"01M7D4ZB4CJBFW5JE67CPH2QRV","web-typography","Select, pair, and implement typefaces for web projects. Use when the user mentions \"font pairing\", \"which typeface\", \"line height\", \"responsive typography\", \"web font loading\", \"type hierarchy\", \"variable fonts\", \"FOUT\u002FFOIT\", \"typographic scale\", or \"the text is hard to read\". Also trigger when choosing between system fonts and web fonts, optimizing font-loading performance, or designing readable long-form content. Covers readability evaluation, CSS implementation, and performance optimization. For overall UI design systems, see refactoring-ui. For dramatic typographic experiences, see top-design.","https:\u002F\u002Fgithub.com\u002Fom-scogo\u002Fskillsh-scraper\u002Fblob\u002Fmain\u002Fdata\u002Fskills\u002Fwondelai-skills\u002Fweb-typography\u002Fweb-typography\u002FSKILL.md",[],[],{"skillNo":91,"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":92,"categories":93,"keywords":94,"publishedAt":19},"01M7D4Z4MVYVY3NMJ18BNV6WNA","https:\u002F\u002Fgithub.com\u002Fom-scogo\u002Fskillsh-scraper\u002Fblob\u002Fmain\u002Fdata\u002Fskills\u002Fwondelai-skills\u002Fweb-typography\u002FSkill.md",[],[]]