By 3d-stories
Generate images using AI models on Replicate (Nano Banana 2, FLUX Schnell) with automatic background removal for transparent PNGs.
Generate images for websites, presentations, and marketing materials using AI models on Replicate (Nano Banana 2, FLUX Schnell) with optional background removal for transparent PNGs. Use this skill whenever the user asks to create, generate, or make images, illustrations, icons, hero images, OG images, logos, favicons, or any visual asset. Also use when implementing issues that require visual assets, or when the user says "generate an image", "create an illustration", "make me an icon", "I need a visual for...", "remove background", "make transparent", or references image generation in any context. If a project needs visual assets and you're about to use placeholder text/emoji instead, use this skill to generate real assets.
Set up the Replicate MCP server for image generation. Detects and installs prerequisites (Node.js, npm, npx), configures the API token, and registers the MCP server with Claude Code. Use when user says "setup replicate", "configure image generation", when generate-image skill reports missing MCP, or when the user gets errors trying to generate images. Also use proactively if you detect Replicate MCP tools are missing when the user asks to generate an image.
A Claude Code plugin for generating images using AI models on Replicate. Supports Nano Banana 2 (high quality) and FLUX Schnell (fast/cheap), with automatic background removal for transparent PNGs.
The plugin provides two skills that instruct Claude how to generate images via the Replicate MCP server. When you ask Claude to create an image, the generate-image skill selects the right model, constructs a prompt, calls Replicate's API through MCP tools, and downloads the result. For transparent images (icons, logos, illustrations), it automatically runs a two-step pipeline: generate on white background, then remove the background.
You → "make an icon of a gold star"
→ Claude (generate-image skill)
→ Replicate MCP (Nano Banana 2)
→ Image generated
→ Replicate MCP (recraft bg removal)
→ Transparent PNG
→ Saved to your project
brew install nodeapt package is outdated)claude plugin install generate-image@generate-image
Then run the setup skill to configure the Replicate MCP server:
/generate-image:setup
The setup skill will:
Ask Claude to generate images naturally:
The skill handles model selection, prompt engineering, background removal, and file management automatically.
| Skill | Description |
|---|---|
/generate-image:generate-image | Generate images — model selection, prompt construction, bg removal, file saving |
/generate-image:setup | Detect and install prerequisites, register MCP, configure API token |
| Model | Best For | Cost | Speed |
|---|---|---|---|
| Nano Banana 2 | Icons, illustrations, complex scenes, text in images | ~$0.04/image | ~20s |
| FLUX Schnell | Quick drafts, bulk generation | ~$0.003/image | ~0.5s |
| Background Remover (recraft) | Transparent PNGs from any image | ~$0.01/image | ~5s |
Nano Banana 2 is the default for production assets. FLUX Schnell is used when cost or speed matters more than quality.
| Problem | Fix |
|---|---|
| "Replicate MCP not detected" | Run /generate-image:setup — it detects and fixes the issue |
| "npx: command not found" in MCP but works in terminal | MCP subprocess PATH issue — re-register with absolute path: claude mcp add -s user replicate -- $(which npx) -y replicate-mcp@latest |
"Failed to connect" in claude mcp list | Token missing or expired — check echo $REPLICATE_API_TOKEN, regenerate at replicate.com/account/api-tokens |
| MCP tools missing after setup | Restart Claude Code — MCP servers load at session start |
| "401 Unauthorized" from Replicate | Bad API token — generate a new one at replicate.com/account/api-tokens |
| Dark halos after bg removal | Source image was on a dark background — regenerate on white |
| nvm-installed Node not found by MCP | nvm only loads in interactive shells — use full path in registration: ~/.nvm/versions/node/v22.x.x/bin/npx |
For detailed troubleshooting, run /generate-image:setup — it diagnoses and fixes most issues automatically.
If you previously used the standalone generate-image skill (installed at ~/.claude/skills/generate-image/), remove it after installing this plugin to avoid duplicate triggering:
rm -r ~/.claude/skills/generate-image/
MIT
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
npx claudepluginhub 3d-stories/generate-image --plugin generate-image11 SDLC workflow skills + 4 workspace management + 1 planning skill + 1 security skill + hooks for Claude Code: project registration, setup, session binding, requirements interviewing, issue creation, feature implementation, bug fixing, refactoring, adversarial review, documentation, dependency updates, security audits, performance optimization, incident response, test suite creation, security pattern syncing, and dangerous pattern blocking with per-project exceptions.
Bidirectional sync between Markdown artifacts and Jira/Confluence Cloud. Push stories and epics to Jira, pull status updates, publish Confluence pages — all from frontmatter-driven .md files.
Persistent virtual display + VNC for headless servers — OAuth, CAPTCHA, 2FA browser interactions.
Comprehensive competitor research pipeline — pulls NotebookLM sources, web research via Google AI Mode, and produces curated market briefs with critique review.
Transcribe audio and video files into organized markdown documents with speaker diarization using AssemblyAI. Includes guided first-time setup, video-to-audio extraction, and configurable output formats.
AI image generation CLI powered by Gemini 3.1 Flash (default) and Gemini 3 Pro. Multi-resolution, aspect ratios, cost tracking, green screen transparency, and reference images.
AI image generation Creative Director powered by Google Gemini Nano Banana models. Claude interprets intent, selects domain expertise, constructs optimized prompts, and orchestrates Gemini for best results.
GPT Image 2 CLI as a Claude Code skill: OpenAI gpt-image-2 + Codex image_generation under one command surface, with masks, transparent backgrounds, custom sizes up to 4K, and structured JSON / JSONL progress output.
Generate and edit images using Gemini's Nano Banana Pro (gemini-3-pro-image-preview) model
Generate optimized prompts for Gemini 2.5 Flash Image (Nano Banana). Use for image generation, crafting photo prompts, art styles, or multi-turn editing workflows with best practices.
Generate images using fal.ai nano-banana model