Minimax Image Generator

Text-to-image and image-to-image generation using MiniMax API. Generates images from text prompts (t2i) or transforms reference images (i2i) using MiniMax's...

Security Scan

The skill's code, instructions, and required credential (MINIMAX_API_KEY) match the stated purpose (text-to-image / image-to-image via MiniMax API) and do not request unrelated secrets or perform unexpected actions.

Purpose & Capability

Name/description claim text-to-image and image-to-image via MiniMax. Declared binary (python3) and env var (MINIMAX_API_KEY) are appropriate. The script posts to api.minimaxi.com and implements the described models and options — requirements and functionality are proportionate to the stated purpose.

Instruction Scope

SKILL.md and the included Python script only: read MINIMAX_API_KEY (env or ~/.openclaw/openclaw.json), build a JSON payload, call the MiniMax API, parse results, optionally download/save image URLs. Instructions do not attempt to read unrelated secrets or system files beyond the single OpenClaw config file used to store the apiKey.

Install Mechanism

No install spec; SKILL.md recommends pip install requests which is standard and minimal. There are no downloads from arbitrary URLs, no extracted archives, and no creation of nonstandard system binaries.

Credentials

Only requires a single API key (MINIMAX_API_KEY) which is the expected credential. The script will also read ~/.openclaw/openclaw.json as an alternate key source (matches SKILL.md). No other tokens, passwords, or unrelated environment variables are requested.

Persistence & Privilege

Skill is not forced always-on and does not modify other skills or system-wide settings. It writes images to a per-skill workspace directory (~/.openclaw/workspace/tmp), which is reasonable for saving outputs.

Assessment

This skill appears coherent and implements what it advertises, but review these practical points before installing: (1) The tool will use your MINIMAX_API_KEY — only provide a key you trust to be used with MiniMax and be aware of billing/quotas. (2) The script will read ~/.openclaw/openclaw.json as an alternate key source; verify that file before use. (3) If you supply image URLs, the script will download them to ~/.openclaw/workspace/tmp — avoid pointing it at URLs you don't trust. (4) The README suggests installing the requests package from PyPI; as with any third-party package, install from the official registry and review versions if you have supply-chain concerns. (5) If you are unsure about trusting the skill source, open and review scripts/minimax_image_gen.py (already included) before running; revoke the API key if you later suspect misuse.

MiniMax Image Generation (Text-to-Image & Image-to-Image)

⚠️ Requires a Coding Plan API Key Subscribe at: https://platform.minimaxi.com/subscribe/coding-plan

Text-to-image (t2i) and image-to-image (i2i) generation tool using MiniMax's image generation API.

---

Setup

1. Configure API Key

openclaw config set skills.entries.minimax-image-generator.apiKey "sk-your-key"

Or add to openclaw.json skills entries:

{
  "skills": {
    "entries": {
      "minimax-image-generator": {
        "apiKey": "sk-your-key"
      }
    }
  }
}

2. Dependencies

pip install requests

---

Architecture

~/.openclaw/workspace/skills/minimax-image-generator/
├── SKILL.md
├── _meta.json
└── scripts/
    └── minimax_image_gen.py

---

Usage

From terminal

# Text-to-image (t2i) - basic
MINIMAX_API_KEY="sk-your-key" python3 scripts/minimax_image_gen.py "a beautiful sunset over ocean"

# Text-to-image with options
MINIMAX_API_KEY="sk-your-key" python3 scripts/minimax_image_gen.py "a cat" --model image-01 --aspect-ratio 16:9 --n 2

# Image-to-image (i2i) - reference image by URL
MINIMAX_API_KEY="sk-your-key" python3 scripts/minimax_image_gen.py "transform to anime style" --image-url "https://example.com/photo.jpg"

# Image-to-image with base64
MINIMAX_API_KEY="sk-your-key" python3 scripts/minimax_image_gen.py "make it brighter" --image-base64 "data:image/jpeg;base64,..."

# Save to file
MINIMAX_API_KEY="sk-your-key" python3 scripts/minimax_image_gen.py "a cat" --save cat.png

From code

from minimax_image_gen import generate_image

# Text-to-image
result = generate_image(
    prompt="a beautiful sunset over ocean",
    model="image-01",
    aspect_ratio="16:9",
    n=1
)

# Image-to-image (URL)
result = generate_image(
    prompt="transform to anime style",
    image_url="https://example.com/photo.jpg"
)

# Image-to-image (base64)
result = generate_image(
    prompt="make it brighter",
    image_base64="data:image/jpeg;base64,..."
)

---

Tool Definition

Name: minimax_image_gen

Input Schema:

{
  "type": "object",
  "properties": {
    "prompt": {
      "type": "string",
      "description": "Image description text, max 1500 chars."
    },
    "model": {
      "type": "string",
      "enum": ["image-01", "image-01-live"],
      "default": "image-01",
      "description": "Model name."
    },
    "image_url": {
      "type": "string",
      "description": "Reference image URL for image-to-image (i2i). Supports public URLs."
    },
    "image_base64": {
      "type": "string",
      "description": "Reference image as base64 Data URL for i2i. Format: data:image/jpeg;base64,..."
    },
    "aspect_ratio": {
      "type": "string",
      "enum": ["1:1", "16:9", "4:3", "3:2", "2:3", "3:4", "9:16", "21:9"],
      "default": "1:1",
      "description": "Image aspect ratio."
    },
    "style_type": {
      "type": "string",
      "enum": ["漫画", "元气", "中世纪", "水彩"],
      "description": "Style (only for image-01-live)."
    },
    "n": {
      "type": "integer",
      "minimum": 1,
      "maximum": 9,
      "default": 1,
      "description": "Number of images to generate."
    },
    "response_format": {
      "type": "string",
      "enum": ["url", "base64"],
      "default": "url",
      "description": "Return format."
    },
    "prompt_optimizer": {
      "type": "boolean",
      "default": false,
      "description": "Enable prompt auto-optimization."
    }
  },
  "required": ["prompt"]
}

Output: JSON with image URLs or base64 data

---

Error Codes

Code	Meaning
0	Success
1002	Rate limit
1004	Auth failed - check API Key
1008	Insufficient balance
1026	Content violation
2013	Parameter error
2049	Invalid API Key