image-cog
Skill8 starsChanged 5 months agoMITView on GitHub
AI image generation and photo editing powered by CellCog. Text-to-image, image-to-image, consistent characters, product photography, reference-based generation, style transfer, sets of images, social media visuals, brand assets, stickers, comics, GIFs. Professional image creation with multiple AI models.
- Reads credentials
- Installs packages
In this repository68 other files it gives its agents.
- animejsSkill
- sketch-animation-videoSkill
- sound-fx-for-videoSkill
- content_series_infographicSkill
- tech-article-seriesSkill
- ansible-playbookSkill
- gin-goSkill
- github-actionsSkill
- iac-hclSkill
- nodejs-nextjs-typescriptSkill
- postgresql-developmentSkill
- programming-paradigmsSkill
- python-shellSkill
- agent-team-harnessSkill
- ai-agent-collaboration-standardSkill
- ci-cd-workflow-specSkill
- config-as-code-specSkill
- execution-ownership-migrationSkill
- feature-lane-release-workflowSkill
- generation-upgrade-acceptanceSkill
- gitops-canonical-resource-deliverySkill
- harness-workflowSkill
- infrastructure-as-code-specSkill
- issue-pr-traceability-standardSkill
- multi-environment-delivery-and-releaseSkill
- project-development-standardSkill
- store-and-startup-homepage-specSkill
- zero-trust-overlay-deliverySkill
- it-infra-continuous-pngSkill
- image-svg-pptx-proSkill
- autoglm-websearchSkill
- blogwatcherSkill
- Browser AutomationSkill
- docxSkill
- find-skillsSkill
- image-resizerSkill
- pdfSkill
- pptxSkill
- Self-Improving Agent (With Self-Reflection)Skill
- skill-security-checkSkill
- skill-vetterSkill
- video-translatorSkill
- wan-image-video-gen-editSkill
- xlsxSkill
- backup-disaster-recovery-and-data-migrationSkill
- capacity-cost-and-resource-lifecycleSkill
- github-actions-operational-dispatchSkill
- incident-response-and-change-managementSkill
- network-dns-tls-edge-managementSkill
- observability-slo-and-alertingSkill
- secrets-identity-and-access-governanceSkill
- service-catalog-and-runbook-standardSkill
- system-db-three-role-securitySkill
- unified-observability-architectureSkill
- vault-eso-tls-certificate-governanceSkill
- ai-tech-news-videoSkill
- ai-ti-wo-shuo-explainerSkill
- it-infra-evolution-videoSkill
- it-infra-evolution-video-v2Skill
- product-intro-videoSkill
- The first 60 shown.
The file
We split this skill into the rules it gives an agent
6 rules. About 0 are common, 4 are distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈4
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 17
Create professional images with AI - from single images to consistent character sets to product photography.
- Distinctive: line 81
Create multiple images of the same character in different scenarios:
- Distinctive: line 113
Use existing images as references for style, character, or composition:
- Distinctive: line 192
For consistency: When creating character series, describe the character in detail first, then reference "the same character" in subsequent …
- Very specific: line 198
Run /cellcog-setup (or /cellcog:cellcog-setup depending on your tool) to install and authenticate.
- Very specific: line 199
OpenClaw users: Run clawhub install cellcog instead.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Image Cog - AI Image Generation Powered by CellCog
Create professional images with AI - from single images to consistent character sets to product photography.
How to Use
For your first CellCog task in a session, read the cellcog skill for the full SDK reference — file handling, chat modes, timeouts, and more.
OpenClaw (fire-and-forget):
result = client.create_chat(
prompt="[your task prompt]",
notify_session_key="agent:main:main",
task_label="my-task",
chat_mode="agent",
)All agents except OpenClaw (blocks until done):
from cellcog import CellCogClient
client = CellCogClient(agent_provider="openclaw|cursor|claude-code|codex|...")
result = client.create_chat(
prompt="[your task prompt]",
task_label="my-task",
chat_mode="agent",
)
print(result["message"])What Models Do We Use
| Model | Provider | Primary Use |
| Nano Banana 2 (Gemini 3.1 Flash Image) | Default image generation — photorealistic scenes, complex compositions, text rendering, multi-turn character consistency | |
| GPT Image 1.5 | OpenAI | Transparent background images — logos, stickers, product cutouts, overlay graphics |
| Recraft | Recraft AI | Scalable vector illustrations (SVG) and icon generation |
Nano Banana 2 is the default model for all image generation. CellCog's agents intelligently route to other models when the task calls for it — for example, transparent PNGs are automatically handled by GPT Image 1.5, and vector/icon requests go to Recraft. If you'd prefer a specific model, just mention it in your prompt (e.g., "use ChatGPT/OpenAI image generation").
What Images You Can Create
Single Image Creation
Generate any image from a text description:
- Scenes: "A cozy coffee shop interior with morning light streaming through windows"
- Portraits: "Professional headshot of a confident woman in business attire"
- Products: "Minimalist product shot of a white sneaker on a marble surface"
- Abstract: "Geometric abstract art in navy and gold"
- Nature: "Misty mountain landscape at sunrise with a lone hiker"
Image Editing
Transform existing images:
- Style Transfer: "Transform this photo into a watercolor painting"
- Background Removal: "Remove the background and place on a clean white backdrop"
- Enhancement: "Enhance the colors and add dramatic lighting"
- Modification: "Change the person's outfit to a red dress"
Consistent Characters
Create multiple images of the same character in different scenarios:
- Character Series: "Create a tech entrepreneur character, then show them: 1) At their desk coding, 2) Presenting to investors, 3) Celebrating a product launch"
- Mascot Variations: "Design a friendly robot mascot, then create versions for: welcome page, error page, success message, loading screen"
- Story Sequences: "Create a main character, then illustrate them in 5 scenes of a journey"
This is powerful for:
- Comic strips and storyboards
- Marketing campaigns with consistent characters
- Video frame generation
- Brand mascots across contexts
Product Photography Style
Professional product visuals:
- Hero Shots: "Product hero shot of a smartwatch on a gradient background"
- Lifestyle Shots: "Smartphone being used by a person in a modern living room"
- Flat Lays: "Flat lay of skincare products with botanical elements"
- 360 Views: "Multiple angles of a leather handbag - front, side, back, detail"
Sets of Related Images
Multiple cohesive images for campaigns or collections:
- Social Media Sets: "5 Instagram post images for a fitness brand - consistent style, varied content"
- Website Heroes: "3 hero images for a SaaS landing page - professional, modern, tech-focused"
- Ad Variations: "4 versions of a product ad with different backgrounds and moods"
- Blog Illustrations: "Set of 6 illustrations for a blog post about productivity tips"
Reference-Based Generation
Use existing images as references for style, character, or composition:
- Style Matching: "Create a new image in the same artistic style as this reference"
- Character Consistency: "Using this person as reference, create a new scene with them hiking"
- Brand Alignment: "Create product images matching this brand's visual style"
- Composition Reference: "Create a similar composition but with different subjects"
Image Specifications
| Aspect | Options |
| Aspect Ratios | 1:1 (square), 16:9, 9:16, 4:3, 3:4, 3:2, 2:3, 21:9 |
| Sizes | 1K (~1024px), 2K (~2048px), 4K (~4096px) |
| Styles | Photorealistic, illustration, watercolor, oil painting, anime, digital art, vector |
| Formats | PNG (default) |
Size recommendations:
- 1K: Quick iterations, thumbnails, social media posts, drafts
- 2K: Standard web content, presentations, marketing materials
- 4K: Hero images, print materials, final deliverables where detail matters
When to Use Agent Team Mode
For image generation, chat_mode="agent team" is recommended for:
- Complex scenes requiring multiple elements
- Consistent character series
- Reference-based generation requiring analysis
- Sets of related images
For simple single images, chat_mode="agent" can work faster.
Example Image Prompts
Professional headshot:
"Create a professional headshot of a friendly Asian woman in her 30s, wearing a navy blazer, soft studio lighting, neutral gray background, confident but approachable expression. 1:1 square, 2K quality, photorealistic."
Product photography:
"Product shot of a premium wireless earbuds case, matte black finish, on a reflective dark surface with subtle blue accent lighting. Minimalist, high-end tech aesthetic. 4:3 landscape, 4K for hero image."
Consistent character set:
"Create a character: young Black male software developer, casual style with glasses, friendly demeanor. Then create 4 images:
- Working at a standing desk with multiple monitors
- In a video call meeting, explaining something
- At a coffee shop with laptop, thinking
- Celebrating with team, high-fiving Keep the character exactly consistent across all images."
Social media set:
"Create 5 Instagram posts for a plant-based meal delivery service:
- Colorful Buddha bowl from above
- Happy person unpacking delivery
- Meal prep containers arranged neatly
- Close-up of fresh ingredients
- Before/after showing ingredients to finished dish Style: bright, fresh, appetizing, consistent warm color grading. 1:1 square format."
Style transfer:
"Transform this uploaded photo of a city street into a Studio Ghibli anime style illustration. Keep the composition and elements but apply the characteristic Ghibli warmth, soft clouds, and whimsical details."
Tips for Better Images
- Be descriptive: "Woman in office" is vague. "Confident woman in her 40s, silver blazer, modern glass-walled office, warm afternoon light" is better.
- Specify style: "Photorealistic", "digital illustration", "watercolor", "minimalist vector".
- Describe lighting: "Soft natural light", "dramatic side lighting", "golden hour glow", "studio lighting".
- Include mood: "Professional and confident", "warm and inviting", "energetic and vibrant".
- Mention composition: "Rule of thirds", "centered symmetry", "close-up", "wide establishing shot".
- For consistency: When creating character series, describe the character in detail first, then reference "the same character" in subsequent prompts.
If CellCog is not installed
Run /cellcog-setup (or /cellcog:cellcog-setup depending on your tool) to install and authenticate.
OpenClaw users: Run clawhub install cellcog instead.
Manual setup: pip install -U cellcog and set CELLCOG_API_KEY. See the cellcog skill for SDK reference.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

