orchesis

<p align="center">
  <img src="docs/assets/banner.png" alt="Orchesis" width="100%">
</p>

<p align="center">
  <strong>See everything your AI agents do.</strong>
</p>

<p align="center">
  Runtime Gateway for AI Agents — block threats, cut token waste, monitor your fleet. One config change.
</p>

<p align="center">
  <a href="https://pypi.org/project/orchesis/"><img src="https://img.shields.io/pypi/v/orchesis?color=a855f7&style=flat-square" alt="PyPI"></a>
  <!-- UPDATE: test count on each release -->
  <a href="https://github.com/poushwell/orchesis/actions"><img src="https://img.shields.io/badge/tests-4%2C912%2B-22c55e?style=flat-square" alt="Tests"></a>
  <a href="https://github.com/poushwell/orchesis/blob/main/LICENSE"><img src="https://img.shields.io/badge/license-MIT-38bdf8?style=flat-square" alt="License"></a>
  <a href="https://www.python.org/"><img src="https://img.shields.io/badge/python-3.10%2B-fb923c?style=flat-square" alt="Python"></a>
  <a href="https://orchesis.ai/scan"><img src="https://img.shields.io/badge/MCP_checks-113-c084fc?style=flat-square" alt="MCP Checks"></a>
</p>

<p align="center">
  <a href="https://orchesis.ai">Website</a> · <a href="https://github.com/poushwell/orchesis/blob/main/QUICK_START.md">Docs</a> · <a href="https://orchesis.ai/scan">MCP Scanner</a> · <a href="https://orchesis.ai/scorecard">Scorecard</a> · <a href="https://orchesis.ai/blog">Blog</a>
</p>

---

## What is Orchesis?

> Your AI agent made 122 API calls. Its built-in loop detector caught zero. Detection was ON. All thresholds configured.
> ([Issue #34574](https://github.com/OpenClaw/OpenClaw/issues/34574))

Orchesis is an open-source HTTP proxy that sits between your AI agents and their LLM providers (OpenAI, Anthropic, Google, Mistral). One config change — set `base_url` to `localhost:8080` — and every request passes through a **17-phase security pipeline**. No SDK integration. No code changes. No vendor lock-in.

- **Security** — Injection detection (96% explicit,