AI Proxy & Token Manager

This skill manages the 'Brain Costs' of the business, ensuring you use the cheapest model that can still do the job.

Core Mechanics (99% Logic)

1. Task Complexity Analysis: Determine if a query needs Claude 3.5 (High reasoning) or can be handled by GPT-4o-mini (Cheaper).
2. Prompt Compression: Strip unnecessary words from prompts to reduce token usage by up to 30%.
3. Usage Tracking: Monitor API credit consumption in real-time across multiple providers.
4. Budget Guardrails: Automatically stop non-essential AI agents if the daily budget is exceeded.
5. API Key Rotation: Swap between different provider keys to take advantage of free tiers or lower-latency regions.

Examples

• "Route all my simple research tasks to the cheapest model to save credits."
• "Summarize my total AI spending across OpenAI and Anthropic this month."