model-audit
Monthly LLM stack audit — compare your current models against latest benchmarks and pricing from OpenRouter. Identifies potential savings, upgrades, and better alternatives by category (reasoning, code, fast, cheap, vision). Use for optimizing AI costs and staying on the frontier.
Install via CLI (Recommended)
clawhub install openclaw/skills/skills/aiwithabidi/model-auditWhat This Skill Does
The model-audit skill is a sophisticated diagnostic tool designed to keep your OpenClaw agent stack performing at peak efficiency. It bridges the gap between your current LLM configuration and the rapidly evolving AI marketplace by tapping into live pricing data from OpenRouter. By analyzing your openclaw.json file, the script categorizes your active models into specialized buckets: reasoning, code, fast, cheap, and vision. It then cross-references this against the latest industry benchmarks to highlight discrepancies where you might be overpaying for performance or missing out on newer, faster, or more capable models that have recently entered the market. The tool calculates potential monthly savings and suggests actionable swaps, ensuring your token budget is always optimized for your specific use cases.
Installation
To integrate this utility, ensure you have the OpenClaw environment initialized. First, set your API credentials by exporting your key: export OPENROUTER_API_KEY='your-key-here'. Install the skill directly from the repository using the terminal command: clawhub install openclaw/skills/skills/aiwithabidi/model-audit. Once installed, you can trigger the audit script by navigating to your skill base directory and running python3 {baseDir}/scripts/model_audit.py. Ensure your openclaw.json is properly configured to allow the script to scan your active model list.
Use Cases
This skill is ideal for AI-driven businesses, software engineers, and automation power users. Use it to conduct monthly cost-benefit analyses of your AI infrastructure, evaluate if a new model release (such as a fresh Gemini or DeepSeek iteration) is ready to replace legacy models in your workflow, or perform deep-dive comparisons between two models to see if the latency/cost difference justifies the switch. It is particularly effective for teams managing high-volume token consumption where small per-token savings scale into significant monthly overhead reductions.
Example Prompts
- "OpenClaw, please run a full model audit and show me how much I could save if I switch my reasoning models to the latest recommended alternatives."
- "Perform a comparison between my current production model claude-3.5-sonnet and gpt-4o, and tell me which is more cost-effective for my code-generation tasks."
- "Identify the best, cheapest model for simple vision tasks in my current stack and list any potential upgrades I should consider."
Tips & Limitations
To get the most accurate audit, ensure your openclaw.json is kept up to date. The script relies on live data from OpenRouter, so a stable internet connection is required during execution. While the audit provides excellent recommendations, always perform a trial run on a non-production workload before swapping models in critical, performance-sensitive infrastructure. Be aware that 'fast' and 'cheap' metrics are based on token costs, so evaluate your specific throughput requirements alongside the model's intelligence level.
Metadata
Not sure this is the right skill?
Describe what you want to build — we'll match you to the best skill from 16,000+ options.
Find the right skillPaste this into your clawhub.json to enable this plugin.
{
"plugins": {
"official-aiwithabidi-model-audit": {
"enabled": true,
"auto_update": true
}
}
}Tags(AI)
Flags: network-access, file-read, external-api
Related Skills
freshsales
Freshsales CRM integration — manage contacts, leads, deals, accounts, tasks, and sales sequences via the Freshsales API. Track deal pipelines, automate lead assignments, log activities, and generate sales reports. Built for AI agents — Python stdlib only, no dependencies. Use for sales CRM, contact management, deal tracking, pipeline reporting, and sales automation.
gemini-video-analyzer
Native video analysis using Google Gemini API. Upload and analyze video files — describe scenes, extract text/UI, answer questions about content, transcribe speech, identify objects and actions. Use when: (1) User sends a video file and wants it analyzed, (2) Video summarization or description needed, (3) Extracting text, UI elements, or information from screen recordings, (4) Answering questions about video content, (5) Comparing multiple videos, (6) Analyzing tutorials, demos, or walkthroughs.
agent-memory
Full AI agent memory stack — Mem0 unified memory engine with vector search (Qdrant) and knowledge graph (Neo4j), plus SQLite for structured data. Complete setup script and tools. Give your OpenClaw agent a real brain with semantic recall, entity relationships, and structured storage.
neon
Neon serverless Postgres — manage projects, branches, databases, roles, endpoints, and compute via the Neon API. Create database branches for development, manage connection endpoints, scale compute, and monitor usage. Built for AI agents — Python stdlib only, zero dependencies. Use for serverless Postgres, database branching, database management, development workflows, and cloud database automation.
onepassword
1Password Connect — vaults, items, secrets management for server-side applications.