AI-Gateway AGENTS.md

A set of instructions for AI coding agents working in the Azure-Samples/AI-Gateway repository. It explains the repository's folders, experiments, and examples involving Azure API Management and AI services.

In plain words
What is it for?
Use it when working on the repository's labs, Jupyter notebooks, infrastructure templates, API policies, or examples for AI agents and model integrations.
Why use it?
It gives an agent the project context needed to find the right files and understand how the repository is organised.

Instructions file for CodexOpenCode

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add instructions/azure-samples/ai-gateway/agents-md
Clone the repo
git clone --depth 1 https://github.com/Azure-Samples/AI-Gateway

Made for: Codex, OpenCode.

Per session 1,322 This file is loaded in full into every session.
When invoked 1,322 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.01322 $0.01322
Opus 5 $0.00661 $0.00661
Sonnet 5 $0.00264 $0.00264
Haiku 4.5 $0.00132 $0.00132

Measured yesterday against content hash 3e5c41e73068, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

AI-Gateway AGENTS.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

AGENTS.md · 118 lines

How it starts

The opening of the file, as written. The whole thing — 118 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AGENTS.md

This document describes the key directories in the AI Gateway workspace for AI coding agents.

Overview

The AI Gateway is an experimental repository exploring the AI Gateway pattern through Azure API Management. It focuses on managing AI services APIs with security, reliability, performance, and cost controls. Labs use Jupyter notebooks with Python, Bicep templates, and Azure API Management policies.


Directory Structure

labs/

Contains hands-on experimental labs, each in its own subdirectory. Labs are structured as Jupyter notebooks with supporting Bicep infrastructure files and APIM policies.

Categories of labs include:

  • AI Agents & MCP: model-context-protocol/, mcp-client-authorization/, mcp-a2a-agents/, mcp-from-api/, mcp-prm-oauth/, mcp-registry-apic/, openai-agents/, ai-agent-service/, realtime-mcp-agents/, gemini-mcp-agents/

  • Model Integration: ai-foundry-sdk/, ai-foundry-deepseek/, ai-foundry-private-mcp/, gemini-models/, aws-bedrock/, slm-self-hosting/

  • Load Balancing & Routing: backend-pool-load-balancing/, backend-pool-load-balancing-tf/, model-routing/

  • Security & Access Control: access-controlling/, content-safety/, private-connectivity/, secure-responses-api/

  • Monitoring & Logging: built-in-logging/, token-metrics-emitting/

  • Rate Limiting & Caching: token-rate-limiting/, semantic-caching/

  • Specialized Features: realtime-audio/, image-generation/, function-calling/, vector-searching/, message-storing/, session-awareness/

  • Operations: finops-framework/, zero-to-production/

  • Lab structure pattern:

  • README.md - README file to describe lab following the standard lab structure.

  • <lab-name>.ipynb - Main Jupyter notebook with step-by-step instructions

  • clean-up-resources.ipynb - Jupyter notebooks used to removed resources when the lab is finished

  • main.bicep - Azure infrastructure deployment template

  • params.json - Temporary file generated automatically for the bicep deployment. This file will not be commited to the repo.

  • *policy.xml - Azure API Management policy files

  • pyproject.toml - Lab-specific Python dependencies (only present when the lab needs deps beyond the root environment).

  • src/ - Supporting source code (when applicable)

Read the full file on GitHub · 118 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. yesterday First seen · 118 lines · 1,322 tokens per session scan A 3e5c41e73068

Subscribe to this mod's changes

AI-Gateway AGENTS.md is an instructions file published in the GitHub repository Azure-Samples/AI-Gateway (980 stars, last pushed 25d ago), licensed MIT. It adds 1,322 tokens to every session, about $0.0066 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.