switchboard
Cost-optimize AI agent operations by routing tasks to appropriate models based on complexity. Use this skill when: (1) deciding which model to use for a task, (2) spawning sub-agents, (3) considering cost efficiency, (4) the current model feels like overkill for the task. Triggers: "model routing", "cost optimization", "which model", "too expensive", "spawn agent", "cheap model", "expensive", "tier 1", "tier 2", "tier 3".
Why use this skill?
Optimize OpenClaw AI costs by dynamically routing tasks to the best-fit models via OpenRouter. Manage budget tiers for your agents.
Install via CLI (Recommended)
clawhub install openclaw/skills/skills/gigabit-eth/routerWhat This Skill Does
The switchboard skill is an intelligent routing engine for OpenClaw that dynamically manages your AI model allocation to maximize cost efficiency without compromising output quality. By leveraging the OpenRouter API, it allows agents to evaluate the complexity of a task and select an appropriate model tier. It effectively categorizes operations into three tiers, ensuring that routine janitorial tasks—such as file reads, status checks, and simple formatting—are handled by cost-effective models, while saving your budget for premium, high-reasoning tasks.
Installation
To enable switchboard, ensure you have an active OpenRouter API key. Add the following to your configuration file at ~/.openclaw/openclaw.json:
{ "openrouter_api_key": "sk-or-v1-..." }
Once the key is configured, you can install the skill by running the command: clawhub install openclaw/skills/skills/gigabit-eth/router. If the installation succeeds, your agent will automatically begin to monitor for routing triggers and cost-optimization requests.
Use Cases
Use the switchboard skill when you are scaling up agent operations or performing repetitive bulk tasks where high-cost model usage would lead to financial bloat. It is ideal for:
- Spawning sub-agents for specialized, low-stakes sub-tasks.
- Selecting models based on context window requirements (e.g., using Gemini 2.0 Flash for massive documentation analysis).
- Budgetary control when running agentic loops that execute hundreds of iterations.
- Balancing speed and intelligence for routine coding tasks.
Example Prompts
- "I'm starting a massive document analysis task. Use the switchboard to route this to a cheap model with a large context window, since deep reasoning isn't required."
- "This task seems too expensive for the current model. Can you spawn a sub-agent using a Tier 1 model to handle the file formatting?"
- "Switch to a more efficient model for these routine status checks; I'm trying to optimize my monthly usage costs."
Tips & Limitations
Always be aware of privacy. Free and Tier 1 models often have different data handling policies compared to premium models. Do not send sensitive authentication credentials or private PII through unmoderated or free-tier models. If you are handling proprietary codebases or sensitive research data, stick to enterprise-grade models in the Tier 2 or Tier 3 categories. Monitor your OpenRouter dashboard regularly to verify that the automated routing is adhering to your intended cost thresholds.
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-gigabit-eth-router": {
"enabled": true,
"auto_update": true
}
}
}Tags(AI)
Flags: external-api