What problem does it solve? Building AI features into JavaScript or TypeScript applications requires handling API authentication, file uploads, streaming responses, and agent orchestration manually. This Skill provides the complete reference for the @inferencesh/sdk package, letting you run 250+ AI models, build agents with tools, and stream results without writing boilerplate integration code. ## Core Features & Use Cases - Run AI Apps: Execute any inference.sh app with typed inputs, fire-and-forget tasks, streaming progress updates, and stateful sessions that keep workers warm. - Agent SDK: Create template or ad-hoc agents with a tool builder API supporting client tools, app tools, sub-agent delegation, webhooks, and human approval workflows. - Frontend Integration: Secure server proxy handlers for Next.js, Express, Hono, Remix, and SvelteKit keep API keys off the browser, with React hooks for chat, streaming, and file uploads. - Use Case: Build a Next.js app where users upload an image, an agent analyzes it with a vision model, and results stream back to the UI in real time. ## Quick Start Install the @inferencesh/sdk npm package and use the JavaScript SDK skill to run an AI app or build an agent in my TypeScript project.
