A performance-focused skill that delegates tasks from OpenClaw to specialized LLMs via the NVIDIA NIM platform.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install openclaw-nim-skill
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 openclaw-nim-skill 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 NVIDIA NIM Integration?
The NVIDIA NIM skill provides a robust bridge between the OpenClaw environment and NVIDIA's high-performance model hosting platform. This integration allows developers to leverage a wide array of state-of-the-art Large Language Models (LLMs) such as GLM-5, Kimi-k2.5, and Llama 3.1. By using this within Openclaw Skills, you can significantly reduce the load on your primary agent's token budget while tapping into specialized reasoning and coding capabilities.
Built for efficiency, this skill enables a modular approach to AI agent development. Instead of relying on a single model for every task, Openclaw Skills can now route specific sub-tasks to the most appropriate NIM-hosted model, ensuring both cost-effectiveness and high-quality outputs across diverse technical domains.
NVIDIA NIM Integration Use Cases
- Offloading complex reasoning tasks to massive models like Llama 3.1 405B.
- Utilizing specialized models like Kimi-k2.5 for long-context or region-specific language tasks.
- Saving primary agent tokens by routing smaller queries to efficient models like Phi-4.
- Integrating diverse AI providers through a unified interface within the Openclaw Skills ecosystem.
How NVIDIA NIM Integration Works
- The OpenClaw agent or a user triggers the skill using a specific model alias and a prompt.
- The skill script identifies the target model configuration from its supported aliases.
- The skill authenticates with the NVIDIA NIM platform using the stored environment variable.
- A request is dispatched to the NVIDIA API to process the prompt through the selected LLM.
- The response is returned to the agent, providing a seamless extension of capabilities for Openclaw Skills.
NVIDIA NIM Integration Setup
- Register an account and obtain an API key at build.nvidia.com.
- Set your environment variable for authentication:
export NVIDIA_API_KEY="your_api_key_here"
- Integrate the command into your Openclaw Skills workflow by adding the following rule to your
CLAUDE.mdfile:
- /nim <model> <prompt>: Call NVIDIA NIM models.
NVIDIA NIM Integration Data Schema & Taxonomy
| Element | Type | Description |
|---|---|---|
| Model Alias | String | A short identifier (e.g., glm5, r1, phi) mapping to a specific NIM endpoint. |
| NVIDIA_API_KEY | Env Var | The required authentication token starting with nvapi-. |
| Scripts | Path | The execution logic resides in scripts/nim_call.py. |
| Prompt | String | The input text sent to the selected NVIDIA NIM model. |
NVIDIA NIM Integration Advanced Features
- Support for a wide range of model aliases including DeepSeek R1 and Microsoft Phi-4.
- Low-latency model switching for dynamic multi-model workflows.
- Easy integration with project-level instructions to extend Openclaw Skills logic.
- Token-efficient delegation to specialized hardware-accelerated endpoints.
SKILL.md
Loading
Related Openclaw Skills
Local Task Runner
A streamlined execution engine for running Node.js scripts and code snippets locally without the overhead of full subagents.
Trakt API Integration
A powerful integration for managing movie and TV show watchlists, history, and ratings via the Trakt.tv API.
Daily Game News Skill
An automated scraping and reporting tool that aggregates news from top gaming websites into structured daily digests.
Self-Improvement Agent Skill
A structured logging and promotion system that enables AI coding agents to learn from errors, user corrections, and discovered best practices.
Tender Offer Arbitrage Scanner
An automated financial analysis tool that scans market data and SEC filings to identify and rank profitable tender offer arbitrage opportunities.
Israeli Stock Analysis
A comprehensive analytical engine for Tel Aviv Stock Exchange (TASE) securities, providing real-time data, fundamental metrics, and technical indicators.
Featured*
