Browse

Understands your task description in plain English and finds matching skills from the catalog, with a 1-sentence explanation of what each does and why it fits.

Converts a wireframe, screenshot, or verbal description into production-ready HTML/CSS/React with attention to typography, spacing, and interactive polish.

Reviews your React or Next.js code against Vercel's performance checklist, flags renders, hydration mismatches, and data-fetching patterns that slow pages down, with refactoring suggestions.

Teaches Claude to navigate websites, fill forms, click buttons, take screenshots, and extract data—turning "go grab my latest analytics" or "fill out this form for me" into one-command tasks.

Asks relentless follow-ups on your design, strategy, or roadmap — one branch at a time — until you've thought through the gaps and can defend every trade-off.

Reads your HTML and CSS, checks for WCAG compliance, semantic markup, color contrast, keyboard navigation, and screen-reader compatibility. Returns a prioritized fix list.

Builds a Docker image, pushes to ACR, deploys to Foundry, runs evals against a dataset, then fine-tunes the model based on failures—all from one command.

Integrates Claude with Azure Search, Speech, Document Intelligence, and OpenAI endpoints. Handles vector search, hybrid queries, transcription, and OCR in one place.

Runs your pre-configured deployment plan — azd up, terraform, container pushes — against Azure. Watches for errors and rolls back on failure.

Connects to AppLens and Azure Monitor to surface logs, resource health, and error traces. Narrows down whether the issue is deployment, CPU, networking, or code.

Builds Bicep/Terraform infrastructure, azure.yaml, and Dockerfiles configured for your stack. Ready to deploy to Azure Container Apps or App Service without manual setup.

Compares Blob, Files, Queues, and Tables; explains when hot vs. cool vs. archive tiers save money; recommends the cheapest fit for your access pattern.