> ## Documentation Index > Fetch the complete documentation index at: https://docs.teamfollowup.ai/llms.txt > Use this file to discover all available pages before exploring further. > ## Agent Instructions > Use the public API base URL: https://api.teamfollowup.ai/api. > Authenticate public API requests with Authorization: Bearer YOUR_API_KEY. > Use product-owned terms: agents, calls, campaigns, contacts, phone numbers, billing, and configuration. # Skills for voice AI > The worry with voice AI is always the same: can you trust it on a live call? Skills are how we make the answer yes. Everyone's first question about voice AI is the same: can I trust it on a real call, with a real customer, when I'm not listening? It's the right question. A voice agent that improvises at the wrong moment costs you the lead. Our answer is **skills**. You already know the idea from tools like Claude Code and Codex: instead of one enormous prompt trying to hold every rule at once, the agent carries focused capabilities and reaches for the right one exactly when it's needed. We brought that to voice. As far as we know, we're the first to do it this way, not as a headline, but as an engineering decision, because it's what makes an agent dependable. Skills change what a voice agent is: * It reaches for the right play at the right moment instead of guessing (**progressive disclosure**). * It drifts and improvises far less, because every moment has its own tight brief (**fewer hallucinations**). * For the things that must be exactly right, it runs real code and returns a fact, not an opinion (**deterministic**). * It gets better over time, because each skill is a small, specific thing we can watch and sharpen, call after call. That last point is the quiet one that matters most. A voice agent built on skills isn't frozen the day it goes live. It's a surface we keep tuning, so it grows more reliable the longer it runs. The full picture: skills you write, tools we ship, and why it all makes for a steadier call.