A specialized skill enabling native MiniMax web search and image analysis within the OpenClaw environment.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install minimax-coding-plan
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install minimax-coding-plan using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is MiniMax Coding Plan?
The MiniMax Coding Plan is a dedicated integration designed to provide agents with direct access to MiniMax-native toolsets. It serves as a high-performance alternative to generic search and vision modules, offering specialized capabilities for web research and multimodal image understanding. By adding this to your repository of Openclaw Skills, you ensure that your agent can handle complex visual data and perform deep web searches using the MiniMax engine's specific logic.
This skill is implemented as a local wrapper script, allowing for seamless execution of tools like web_search and understand_image. It is particularly effective when the standard media stack is insufficient or when the user requires the specific accuracy associated with MiniMax models for vision-based tasks.
MiniMax Coding Plan Use Cases
- Performing deep web research using the native MiniMax search engine for high-relevance results.
- Analyzing UI/UX screenshots to describe layouts or extract component logic for frontend development.
- Extracting text and metadata from various image formats including JPEG, PNG, and WebP.
- Acting as a reliable fallback for image understanding when the primary OpenClaw media route is unavailable.
How MiniMax Coding Plan Works
- The agent invokes the local shell wrapper located at the project base directory.
- The system determines the specific tool to execute based on the command arguments: either web_search or understand_image.
- The skill manages authentication by checking for a MINIMAX_API_KEY or scanning predefined OpenClaw directories for OAuth profiles.
- For vision tasks, the script processes local file paths or remote URLs and sends them to the MiniMax API.
- The API returns a structured JSON response which the skill parses, providing the agent with the requested analysis or search data.
MiniMax Coding Plan Setup
To use this skill within your collection of Openclaw Skills, ensure you have a valid MiniMax API key or OAuth profile configured.
# Example: Setting the API key environment variable
export MINIMAX_API_KEY='your_secret_key_here'
# Example: Running a web search via the local script
bash ./scripts/minimax-plan.sh web_search --query "MiniMax M2.5 release note"
MiniMax Coding Plan Data Schema & Taxonomy
The skill utilizes a structured JSON communication format for both input parameters and output responses.
| Tool | Parameter | Description |
|---|---|---|
| web_search | --query | The search string to be processed by the MiniMax engine. |
| understand_image | --prompt | The natural language instruction for the vision model. |
| understand_image | --image-source | The local file path or web URL of the image. |
Response data includes a base_resp object for status checking and a content field containing the primary output from the MiniMax model.
MiniMax Coding Plan Advanced Features
- Multi-source image processing supporting both local filesystem paths and remote HTTPS URLs.
- Intelligent authentication discovery across multiple locations including OPENCLAW_AGENT_DIR and ~/.openclaw.
- Robust error handling that exposes native MiniMax API status codes for debugging.
- Broad compatibility with modern image formats like WebP alongside standard JPEG and PNG files.
SKILL.md
Loading
Related Openclaw Skills
Douyin Upload Skill
A professional CLI tool to automate Douyin video uploads, metadata generation, and AI-driven captioning.
Yuc's Anime List Scraper
A specialized scraper for extracting, validating, and organizing quarterly anime lineup data from yuc.wiki into structured formats.
AnimeGarden
AnimeGarden is a specialized resource discovery skill that aggregates anime torrent data from multiple mirrors like dmhy and bangumi.moe into a unified API.
Yida Dev Assistant
A specialized tool for DingTalk Yida low-code development, facilitating form creation, JS action scripting, and workflow automation.
File Writer
A specialized strategy for AI agents to safely write and modify large files without risking data truncation or corruption.
File Writer
A robust strategy for AI agents to safely write and update large files without risk of truncation or data loss.
Featured*
