Skill UI

Master building robust AI workflows with Anthropic's core APIs. This guide details two powerful patterns: real-time streaming via SSE events for interactive chat UIs, and the cost-efficient Message Batches API for asynchronously processing up to 100,000 requests. Includes comprehensive code examples for Python and TypeScript.

This guide provides comprehensive strategies to significantly reduce spending on the Anthropic Claude API. Key techniques include model routing (selecting the most cost-effective model for the task), prompt caching for repeated prompts, using Message Batches for asynchronous tasks, and implementing real-time spend tracking and budget alerts. Essential for managing large-scale, high-cost AI deployments and optimizing cloud LLM expenditures.

This guide teaches developers how to implement robust, event-driven architecture patterns with the Anthropic Claude API. It covers two key methods: Server-Sent Events (SSE) for real-time streaming responses, and the Message Batches API for high-volume, asynchronous bulk processing. Use this when building real-time chat features or processing large datasets offline.

Implement robust handling for AssemblyAI's asynchronous transcription completion events via webhooks. This guides users on how to set up secure endpoints that receive POST requests upon job completion or failure. It covers submitting transcripts with webhook URLs, verifying authentication headers, and processing the final transcript data, including running subsequent analyses like summarization.

Utilize Anthropic Message Batches API for high-volume, asynchronous bulk processing of AI tasks. This method allows processing up to 10,000 requests per batch, ideal for summarizing large document sets or running repetitive tasks where immediate, synchronous results are not required. The process involves creating the batch, polling for status completion (within 24 hours SLA), and then retrieving detailed results, including success and error statuses for each item.

This guide provides a robust implementation for handling asynchronous webhooks from ElevenLabs. It covers the critical steps of receiving POST notifications (such as transcription completion, post-call audio, and failure events), performing secure HMAC-SHA256 signature verification, and routing the event data to specific processing handlers. Ideal for integrating ElevenLabs into conversational AI pipelines, ensuring data integrity and reliable background processing.

A comprehensive skill for generating high-quality AI images and videos using advanced models like Seedance 2.0, Kling 3.0, and Midjourney. It supports diverse workflows including text-to-image, image-to-video, and video continuation. The system handles asynchronous generation, robust delivery, and multi-channel integration (e.g., Feishu, Slack), making it ideal for creative content production and automated media workflows.

This skill provides implementation patterns for interacting with RunwayML webhooks. It enables complex tasks such as AI video generation, image editing, and advanced creative content workflows. By using the provided Python SDK pattern, developers can trigger video creation tasks (e.g., image-to-video) and handle the asynchronous results, making it ideal for integrating high-end creative AI into applications.

A comprehensive reference architecture for building scalable, cost-effective, and robust AI services. This system manages the full AI lifecycle, including intelligent model routing (for quality/cost trade-offs), response caching (using Redis), asynchronous batch processing, and fine-tuning pipeline management via Together AI's OpenAI-compatible API. Ideal for enterprise applications requiring high availability across multiple open-source models.

This utility handles asynchronous webhook events from Together AI, enabling developers to build robust, event-driven Machine Learning pipelines. Instead of constantly polling the API for status updates (e.g., fine-tuning completion, batch job status, model deprecation), you can subscribe to these webhooks. It includes logic for signature verification, event routing, and idempotency, making it ideal for mission-critical AI workflows.

This skill provides access to the Anthropic Claude Managed Agent runtime, which is a cloud-native, persistent environment for running complex, long-running, or asynchronous AI tasks. Unlike local WASM agents, the managed runtime provides a full container, persistent filesystem, and robust state management across multiple turns. It supports the entire agent lifecycle: creation, prompting, status inspection, event logging, and mandatory termination.

An adapter for generating short, browser-playable video assets using OpenRouter's asynchronous video generation API. This skill is designed for meta-skills to integrate dynamic multimedia content into web pages. It handles the full job lifecycle—submitting the request, polling for completion, and downloading the final MP4 asset—and adheres strictly to the required manifest reporting format.