# Voxgig Agent Experience

> Agent Experience is your API as an AI agent sees it. An MCP server, tool schemas an agent can reason about, and error semantics that fail safely, generated from the same spec as your SDK.

## What agent-ready actually means

- **The MCP server is the front door** for agentic callers. Build it on purpose, generated from the same OpenAPI spec as your SDK so the two cannot disagree.
- **Tool schemas an agent can reason about.** Most first attempts are useless to agents because the descriptions were written for humans skimming docs, not for a machine deciding what to call.
- **Error semantics that fail safely.** A vague 500 gets retried forever. A 429 with no Retry-After retries at the wrong rate. Agents need more care here, not less.
- **Rate limits tuned for autonomous traffic.** Agent traffic is bursty, repetitive, and prone to tight retry loops.
- **Safety guardrails.** Some operations have no business running without a human saying yes first. Deciding which, and tagging them, is real work.

## How we engage

- **MCP foundation sprint**, fixed-price, 4 to 6 weeks. Your first MCP server end to end, tested against Claude, ChatGPT and Cursor before you hand it to a customer.
- **AX audit**, fixed-price, 2 weeks. We test your MCP server against the major agentic clients, score it, and walk you through every finding.
- **AX retainer**, monthly. Tool schema reviews, customer escalations where an agent failed, roadmap reviews.
- **Joint DX and AX engagement.** One spec drives both the SDK and the MCP server, so the work overlaps.

## See it in practice

Voxgig runs its own MCP server at https://voxgig.com/mcp, over Streamable HTTP. Details at https://voxgig.com/developers.

## Related

- [SDK Generator](https://voxgig.com/sdk)
- [Developer Experience](https://voxgig.com/api-experience/developer-experience)
- [Contact](https://voxgig.com/contact)