OnDeckLLM for Openclaw

OnDeckLLM is a localhost dashboard for managing LLM provider health, routing priority, and cost tracking.

canonflip-git
v1.4.3
Mar 23, 2026
0
705
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install ondeckllm

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install ondeckllm 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 OnDeckLLM?

OnDeckLLM serves as a sophisticated AI Model Lineup Manager designed to streamline how developers interact with multiple LLM providers. By providing a centralized localhost web dashboard, it enables users to orchestrate complex model routing and establish robust batting-order fallback chains. Whether you are balancing local Ollama instances with cloud providers like Anthropic or OpenAI, this tool ensures your AI workflows remain resilient and cost-effective.

As a core component within the ecosystem of Openclaw Skills, OnDeckLLM automates the discovery of providers from existing configurations. It bridges the gap between raw JSON configuration files and an intuitive visual interface, allowing for real-time health checks, latency testing, and one-click synchronization of model lineups. It is the essential command center for anyone serious about managing a diverse AI model stack.

OnDeckLLM Use Cases

  • Establishing automated fallback chains (batting orders) to ensure high availability when primary LLM providers fail.
  • Monitoring provider health and latency to optimize response times for AI-driven applications.
  • Tracking and visualizing session costs using JSONL logs and integrated Chart.js dashboards.
  • Managing local and remote Ollama instances alongside major cloud providers like Groq, Mistral, and DeepSeek.
  • Synchronizing provider configurations across multiple Openclaw Skills environments.

How OnDeckLLM Works

  1. The tool auto-discovers existing LLM provider configurations from the standard Openclaw directory.
  2. Users access the localhost web interface to rank providers in a preferred batting order.
  3. The dashboard performs active health checks and latency pings to verify provider status.
  4. Configuration changes, such as adding new API keys or adjusting model priorities, are synced back to the global config file.
  5. Usage data is captured in a local JSONL log for cost analysis and reporting.

OnDeckLLM Setup

Install the package globally using npm to integrate it into your Openclaw Skills toolkit:

npm install -g ondeckllm

To launch the dashboard on the default port (3900):

ondeckllm

To run on a custom port:

PORT=3901 ondeckllm

OnDeckLLM Data Schema & Taxonomy

OnDeckLLM organizes its data within the ~/.ondeckllm/ directory using the following structure:

File/Folder Description
config.json Stores provider settings, dashboard port, and remote Ollama URLs.
usage.jsonl An append-only log tracking token usage and estimated costs per session.
profiles/ A directory containing saved batting-order profiles for different project needs.
~/.openclaw/openclaw.json The primary external configuration file synced by the tool.

OnDeckLLM Advanced Features

  • Multi-provider support including xAI/Grok, OpenRouter, and Together AI.
  • One-click sync that pushes local dashboard changes directly to Openclaw Skills configuration files.
  • Remote Ollama integration for managing GPU clusters over a local network.
  • Programmatic status checking via helper scripts for background service monitoring.
  • Built-in cost tracking with visual data representation using Chart.js.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*