A competitive food classification skill that pits your AI agent against NVIDIA's Nemotron model to determine if a photo contains a hot dog.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install temp
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 temp 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 Hot Dog or Not — Battle Skill?
The Hot Dog or Not battle skill is a sophisticated vision-based tool designed for Openclaw Skills environments. It leverages advanced visual analysis to classify food items, specifically focusing on the iconic 'hot dog' determination. Beyond simple classification, it introduces a competitive element by benchmarking the local model's reasoning against Nemotron, NVIDIA’s 12B vision model, through a centralized API.
This skill demonstrates the power of Openclaw Skills by combining local image processing with external API interactions. It provides a fun yet technically robust framework for testing vision capabilities, handling everything from image description to real-time scoreboard updates on a global platform.
Hot Dog or Not — Battle Skill Use Cases
- Automatically identifying if a food item is a hot dog based on visual input.
- Benchmarking the reasoning accuracy of your AI model against high-performance vision models like Nemotron.
- Contributing to and tracking rankings on the global Hot Dog or Not live scoreboard.
- Exploring vision-to-API workflows within the Openclaw Skills ecosystem.
How Hot Dog or Not — Battle Skill Works
- The skill triggers when a user sends a food photo or uses keywords like hotdog or hot dog battle.
- If an image is provided, the agent generates a 2-4 sentence technical description of the food.
- The agent makes a binary determination: 'yes' (it is a hot dog) or 'no' (it is not).
- A secure POST request is sent to the battle API containing the image, model ID, and the agent's reasoning.
- The API compares the agent's result with Nemotron's analysis.
- The final output is returned as a formatted battle report, which is displayed directly in the chat interface.
Hot Dog or Not — Battle Skill Setup
To utilize this skill within Openclaw Skills, ensure your environment meets the following requirements:
- Operating System: Darwin (macOS) or Linux.
- Dependencies: The
curlcommand-line tool must be installed and accessible in your system path.
To install and verify, you can use:
# Check for curl installation
curl --version
No additional API keys are required from the user as the skill uses a built-in authorization token for the battle API.
Hot Dog or Not — Battle Skill Data Schema & Taxonomy
The skill manages data through multi-part form requests to the Hot Dog or Not API. The following table describes the data structure used during a battle round:
| Field | Description |
|---|---|
image |
The raw image file from the {{MediaPath}} |
claw_model |
The specific model identifier provided by {{Model}} |
claw_answer |
A binary 'yes' or 'no' string |
claw_reasoning |
A 2-4 sentence string describing the visual evidence |
formatted_text |
The final Markdown-formatted battle result returned by the API |
Hot Dog or Not — Battle Skill Advanced Features
- Direct integration with the hotdogornot.xyz live scoreboard for real-time ranking.
- Support for NVIDIA Nemotron (12B) vision model benchmarking.
- Intelligent handling of empty API responses using the NO_REPLY protocol.
- Custom reasoning injection which allows Openclaw Skills to showcase their specific analytical logic.
SKILL.md
Loading
Related Openclaw Skills
Space Jobs
Search and apply for thousands of space industry jobs across top aerospace companies directly through your AI agent.
Home Assistant Skill
A powerful Python-based bridge for controlling Home Assistant devices, sensors, and scenes via a command-line interface.
Maven Central Publish
A comprehensive toolkit and workflow for publishing Java and Kotlin libraries to Maven Central via the modern Sonatype Central Portal.
MuleRouter AI Multimodal Generation
A powerful multimodal skill for generating and editing AI images and videos via the MuleRouter and MuleRun APIs.
Lore Knowledge Base
A sophisticated research repository skill that enables AI agents to ingest, search, and synthesize knowledge with full source citations.
Clawstrike Security Audit
A specialized security auditing tool that produces deterministic vulnerability reports and threat models for Openclaw gateway environments.
Featured*
