Cloudflare Skills
cloudflare/skillsCloudflare's official 19-skill pack for building on Workers, Durable Objects, the Agents SDK, Cloudflare One and more, with an optional bundled MCP server.
Recommended
Check first Depth 4/5
Azure AI Gateway (Azure Skills) skill
Microsoft skill for configuring Azure API Management as an AI gateway: token limits, semantic caching, content safety and MCP rate limiting.
Our verdict Recommended
Useful for teams already on APIM; the policy reference is specific, but the caching claims in the file are the vendor's own.
What the azure-aigateway skill does
azure-aigateway helps an agent configure Azure API Management (APIM) as a gateway in front of AI models, MCP tools and agents. SKILL.md groups tasks into model governance (token limits, semantic caching, load balancing, token usage tracking), tool governance (rate limiting MCP servers, converting an OpenAPI API into an MCP server) and agent governance (content safety, jailbreak detection), plus configuration of Azure OpenAI or AI Foundry backends and testing the gateway. A quick-reference table maps policies such as azure-openai-token-limit, the semantic cache lookup and store pair, azure-openai-emit-token-metric, llm-content-safety and rate-limit-by-key to their purpose. It shows az apim commands to get the gateway URL, list backends, list subscription keys and send a test chat-completions request with curl. Reference files cover policies, patterns, troubleshooting and auth best practices; there are nine files and no scripts. Deploying APIM itself is left to azure-prepare, and the frontmatter says it requires the Azure CLI.
Good for
- Putting token limits, caching and content filtering in front of Azure OpenAI deployments
- Exposing or protecting MCP tools behind APIM
Not for
- Creating the APIM instance (use azure-prepare)
- Gateways on other platforms
Depth 3/5
5 KB SKILL.md plus eight reference files on policies, patterns and troubleshooting; no scripts or evals.
Dependencies and costs
- Azure account or subscription
- Azure CLI signed in with az login
- Azure MCP Server (installed with the plugin; needs Node.js 18+ for npx)
- An Azure API Management instance
- Azure OpenAI or AI Foundry model deployments to put behind it
- Telemetry: The skill files contain no telemetry code. The Azure MCP Server that the plugin wires in has its own telemetry; the repo README says to disable it with AZURE_MCP_COLLECT_TELEMETRY=false.
Security review
Review before installingDocumentation-only skill (9 files) for setting up Azure API Management as an AI gateway: semantic caching, token limits, content safety, load balancing. It changes APIM resources through az and sends test requests to your own gateway with its subscription key; it imports one public OpenAPI spec from GitHub.
Worth knowing
- Relies on a CLI or package we could not reviewUses az (apim, cognitiveservices) and Azure MCP tools to configure API Management as an AI gateway..github/plugins/azure-skills/skills/azure-aigateway/references/patterns.md:23
- Makes network requestsTest step curls your own APIM gateway with an Ocp-Apim-Subscription-Key header; patterns.md line 71 imports an OpenAPI spec from raw.githubusercontent.com (Azure/azure-rest-api-specs) into your APIM..github/plugins/azure-skills/skills/azure-aigateway/SKILL.md:63
- Installs packages or toolsSDK docs tell you to pip install or npm install Azure packages..github/plugins/azure-skills/skills/azure-aigateway/references/sdk/azure-ai-contentsafety-py.md:8
What we checked
The skill folder (9 files) on Oct 29, 2026. Not a full review: it relies on code outside the repository: Azure CLI (az) and the az extensions it installs on demand; not part of the files we have; Azure MCP Server, started by the plugin through npx; its configuration sits outside this skill folder; Azure SDK packages named in the reference docs.
NVIDIA SkillSpector static scan: 3 pattern hits (Data Exfiltration 3). Three medium data-exfiltration hits are curl examples to the user's own gateway URL with a subscription key placeholder; the credential matches are best-practice text.
A review, not a certification. Skills change after we check them — read the files before installing.
License
MIT (declared in the SKILL.md frontmatter); the repo is a mirror synced from microsoft/GitHub-Copilot-for-Azure and has no top-level LICENSE file in our copy. Commercial use allowed under MIT.
Can it run unattended?
Not really. Policy changes on a live gateway need review, and the skill has no loop or verification harness.
Alternatives
Google Agent Skills
google/skillsGoogle's official collection of about 180 skills for Google Cloud, Gemini APIs, Analytics, Data Manager, sign-in and other Google products.
Recommended
Broad access Depth 3/5
Vercel Agent Skills
vercel-labs/agent-skillsVercel's official 9-skill pack: React and Next.js performance rules, composition patterns, UI and writing review, React Native, deploy and cost audit.
Recommended
Broad access Depth 3/5
FAQ
What does the azure-aigateway skill do?
It guides configuring Azure API Management as an AI gateway, with policies for token limits, semantic caching, token metrics, content safety and rate limiting, and with backends for Azure OpenAI and Foundry models.
Does azure-aigateway deploy API Management?
No. It tells you to use azure-prepare to deploy APIM and focuses on configuring policies and backends on an existing instance.
Does the skill handle API keys?
Its examples list APIM subscription keys with az and use one in a test request, so the key can appear in the agent session. Use a test subscription key and rotate it if it leaks into logs.
How do I install azure-aigateway?
Install Microsoft's Azure plugin: /plugin install azure@claude-plugins-official in Claude Code, or add microsoft/azure-skills to the Codex marketplace.
Sources: GitHub repository; install counts from skills.sh via Best Skills (CC BY 4.0). Review based on the published SKILL.md.