Azure AI Gateway is an Azure API Management tier that gives applications one governed way to call AI model APIs and MCP tools while platform teams apply security, policy, networking, and observability controls. It is used by developers and platform teams working with models and tools from multiple providers, and the repository's coding-agent plugin discovers registered models and MCP servers to help build applications on an existing gateway.
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.
npx skills add Azure/ai-gateway --skill use-ai-gatewaygit clone --depth 1 https://github.com/Azure/ai-gatewayWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/azure/ai-gateway/use-ai-gateway)<a href="https://agentmods.dev/skills/azure/ai-gateway/use-ai-gateway"><img src="https://agentmods.dev/badge/skills/azure/ai-gateway/use-ai-gateway/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/azure/ai-gateway/use-ai-gateway"><img src="https://agentmods.dev/badge/skills/azure/ai-gateway/use-ai-gateway.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00089 | $0.08568 |
| Opus 5 | $0.00044 | $0.04284 |
| Sonnet 5 | $0.00018 | $0.01714 |
| Haiku 4.5 | $0.00009 | $0.00857 |
Grade C, and why
use-ai-gateway scanned grade C with 3 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 10d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
- **OpenAI Codex** — add to `~/.codex/config.toml` (or a trusted project `.codex/config.toml`), passing the key by env-var reference so the secret isn't written into config: Reads MCP configurationmediumAgent snooping
mcp.json carries server URLs and auth tokens; reading it lets a mod discover and abuse other integrations.
- **Claude Code** — the `mcpServers` JSON above is exactly the shape of a project-scoped `.mcp.json`. Claude Code expands `${AI_GATEWAY_API_KEY}` from the environment in HTTP `headers`. Run `claude mcp list` (or `/mcp`) Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
HOST=$(curl -s "${AUTH[@]}" "$B?api-version=$VER" | jq -r '.properties.gatewayUrl') How it starts
The opening of the file, as written. The whole thing — 534 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Use an AI Gateway (models & MCP tools)
Invoked as
/ai-gateway:use-ai-gatewaywhen installed via the plugin. It is also invoked automatically by the/ai-gateway:discoverand/ai-gateway:buildcommands.
This skill is coding-agent–agnostic: any assistant that can read this file and run shell commands can follow it.
Use this skill when a developer wants to use the models and/or MCP tools registered in an existing AI Gateway from their application — whatever they are building (a web app, backend service, script, notebook, data pipeline, or a standalone agent). This single skill takes the user all the way from discovery to working, integrated code:
- Discover the models and MCP tools registered in the gateway.
- Select the ones that match the user's intent (and confirm with them).
- Retrieve the credential needed to call them.
- Integrate them into the user's application, choosing the path that fits what they're building:
- Call a model over the gateway's OpenAI-compatible passthrough (any language, framework, or raw HTTP).
- Connect MCP tools from an MCP-capable client/app.
- Scaffold a standalone agent (GitHub Copilot SDK) if the user wants a ready-to-run agent project.
Scope (consumption only)
This skill is strictly for developers consuming an existing AI Gateway. It is read-only against the gateway and only generates client code that calls already-registered models and tools.
- ✅ Allowed: list models, list tools/MCP servers, read an existing API key, generate integration/client/agent code.
- ❌ Not allowed: creating, provisioning, updating, or deleting gateways, models, tools, connections, products, or any other resource. Never issue ARM
PUT/PATCH/DELETEcalls. If a user asks to create or provision a gateway or model, tell them that is an administrator task done in the AI Gateway Portal and is out of scope for this skill.
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.
- 10d ago First seen · 534 lines · 89 tokens per session scan C c7cf6e506c2c
use-ai-gateway is a skill published in the GitHub repository Azure/ai-gateway (3 stars, last pushed 14d ago), licensed MIT. It adds 89 tokens to every session and 8,568 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it C with 3 findings (reads agent configuration directories, reads mcp configuration, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
agent-platform-rag-engine-management
Manage and query Agent Platform RAG Engine Corpora and retrieve grounded contexts using the Google GenAI SDK. Use when listing RAG corpora or files, inspecting a corpus, retrieving contexts, or generating content grounded in a RAG corpus. Do not use for standard database queries (use SQL/Spanner skills), Google…
agent-platform-model-registry
Agent Platform Model Registry Management. Use when you need to upload, list, describe, update, or delete machine learning models (and their versions) in the Agent Platform Model Registry. Don't use for model training, model deployment to endpoints, or managing non-Agent Platform models.
foundry-config-setup
Resolve missing setup caused by a hardcoded Foundry project endpoint or model in a sample. Use when a sample fails because it uses a placeholder/hardcoded projectendpoint (for example "https://your-project.services.ai.azure.com") or a hardcoded model instead of reading them from the environment.
google-cloud-solution-agentic-analytics-spark-knowledge-catalog
Discovers requirements and generates guidance to design and deploy a governed, secure agentic-analytics solution for data that's distributed across Google Cloud, other cloud providers, or on-premises. Data that's outside Google Cloud (such as data from Databricks, Snowflake, Salesforce, SAP, or Oracle systems) is…
training-check
Interactively monitor training metrics from the current Codex session, periodically checking WandB or fallback logs for NaN, divergence, plateaus, and broken runs.
nemo-automodel-launcher-config
Configure NeMo AutoModel job launches for interactive runs, Slurm clusters, and SkyPilot cloud execution.