A decentralized MCP server providing anonymous, pay-per-request LLM inference via Lightning Network micropayments.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install mcp-llm-inference
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 mcp-llm-inference 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 LLM Inference (L402)?
This skill enables AI agents to access high-quality Large Language Models without the friction of traditional API key management or account registration. By leveraging the L402 protocol, it facilitates anonymous inference through micropayments on the Lightning Network, making it one of the most privacy-centric Openclaw Skills available today.
Instead of managing monthly subscriptions or sharing personal information, developers can empower their agents to pay exactly for what they use. This is ideal for building decentralized applications or autonomous agents that need to handle their own resource consumption and billing cycles without human intervention.
LLM Inference (L402) Use Cases
- Agents requiring LLM access without static API key management.
- Privacy-conscious workflows that demand anonymous inference without KYC requirements.
- Autonomous agent-to-agent service consumption using Lightning Network sats.
- On-demand text generation and chat completions without billing portal overhead.
How LLM Inference (L402) Works
- The agent queries available models and current pricing using the model discovery tool.
- A request is initiated for either chat completion or single-prompt text generation.
- The L402 protocol triggers a Lightning Network invoice specifically for the requested inference.
- Once the micropayment (typically around 10 sats) is settled, the model response is instantly delivered.
- The session can maintain message history for multi-turn conversations through standard tool parameters.
LLM Inference (L402) Setup
To integrate this into your environment, add the server configuration to your MCP settings file:
{
"mcpServers": {
"llm-inference": {
"command": "npx",
"args": ["-y", "@vbotholemu/mcp-llm-inference"],
"env": {
"L402_API_BASE_URL": "https://api.nautdev.com"
}
}
}
}
LLM Inference (L402) Data Schema & Taxonomy
The skill utilizes the following tool schema to interact with the inference provider:
| Tool Name | Purpose | Key Parameters |
|---|---|---|
| chat_completion | Multi-turn dialogue | messages, model |
| generate_text | Single prompt output | prompt, model |
| list_models | Discovery | N/A |
All interactions are logged within the MCP host, while the financial transactions are handled externally via the Lightning Network, ensuring this remains a lightweight addition to your collection of Openclaw Skills.
LLM Inference (L402) Advanced Features
- Native L402 protocol support for programmatic micropayments.
- Anonymous model discovery and dynamic pricing updates.
- Support for multi-turn chat history within the chat_completion tool.
- Zero-config deployment via npx for rapid agent prototyping.
- Compatible with any Lightning-enabled wallet for automated settlement.
SKILL.md
Loading
Related Openclaw Skills
ORM Weather Routing Nav Voyage Distance
A professional maritime navigation tool that leverages the NavOptima system to calculate high-precision port-to-port distances and voyage timelines.
Dependency Guard
A security-first skill for AI agents to audit and guard software dependencies using Socket supply-chain intelligence.
Website Manager
A comprehensive automation skill for building, redesigning, and managing websites powered by Notion and deployed via Netlify.
Blueprinter
A high-precision tool for creating flat, engineering-style technical diagrams and system architectures using standard web technologies.
AI CLI Architecture & Engineering Strategy
A comprehensive engineering framework for designing, developing, and iterating robust AI command-line interfaces.
Deep Research Agent
A sophisticated research skill that decomposes complex queries into multi-phase workflows to produce comprehensive, well-cited technical reports.
Featured*
