---
title: "Xedant Proxy Agent — Product Launch"
id: "2216"
type: "post"
slug: "proxy-agent-launch"
published_at: "2026-10-06T09:00:00+00:00"
modified_at: "2026-10-19T22:48:16+00:00"
url: "https://xedant.com/blog/proxy-agent-launch/"
markdown_url: "https://xedant.com/blog/proxy-agent-launch.md"
excerpt: "Today Xedant Proxy Agent launches — one address for all your AI models. Your programs…"
taxonomy_category:
  - "Build in Public"
---

# Xedant Proxy Agent — Product Launch

Today **[Xedant Proxy Agent](/agents/proxy)** launches — one address for all your AI models. Your programs do not talk to a dozen vendors separately, but to one address on your own server.

Vendor keys live in one place and never spread across employee computers. Every person or contractor gets their own access key with their own spend limits and allowed models, and the spend is visible in money, not in tokens. It is a standalone application on your server: it serves all your programs at once, not only your agent.

### Key capabilities

One address instead of ten: programs talk to the proxy, not to every vendor directly

OpenAI, Anthropic and DeepSeek behind one address, whichever program you use

Vendor keys live on your server and never spread across employee computers

Your own access keys for teams and contractors, with revocation and one-click regeneration

Spend limits per key — per day, per week, per month and in total — counted in dollars

Format conversion: a program written for one vendor works with another vendor’s model

Secret hiding: passwords and keys in requests are replaced with marks and put back afterwards

Ordinary traffic too: HTTP and SOCKS5 on one listener, with its own port and its own login requirement on every channel

Every request leaves a log entry, and a chat with the AI agent is built right into the interface. Details and documentation are on the product page: [Xedant Proxy Agent](/agents/proxy)
.