Skill UI

Defuddle is a CLI tool to extract clean Markdown from HTML web pages. It removes navigation, ads, and clutter, reducing token usage. Users can output Markdown, JSON, or HTML, and extract metadata. Preferred over WebFetch for standard pages, it provides cleaner content for processing and analysis.

Automatically analyzes CI/CD performance metrics, compares them against baselines and thresholds, and flags statistical regressions so teams catch response-time or throughput degradations before deployment.

Performs security-focused differential review of PRs, commits, and diffs, adapting depth to repo size, tracing git history, estimating blast radius, checking coverage, and producing detailed markdown reports to stop regressions.

Provides advanced patterns to significantly boost the performance and scalability of Documenso integrations. Key techniques covered include using templates to drastically reduce API round-trips, implementing metadata caching (e.g., using Redis) to minimize database lookups, and utilizing batch processing with concurrency control (p-queue) or asynchronous job queues (Bull) for handling large volumes of documents efficiently.

This skill provides patterns for building highly resilient applications that interact with external APIs. It covers crucial concepts like Exponential Backoff with Jitter, request queuing for bulk operations, and the Circuit Breaker pattern. Use this when your application needs to handle rate limit errors (429), transient network failures, or temporary service outages, ensuring reliable throughput and graceful degradation.

Automates Dropbox file, folder, sharing, upload/download, search and metadata workflows via the Rube MCP (Composio) toolkit while always querying the latest schemas before invoking storage tools.

A comprehensive workflow for implementing advanced search and retrieval features within Evernote. It allows users to build complex queries using search grammar, filter notes by tags, notebooks, dates, and completion status, and paginates through large result sets. The service also enriches results by mapping GUIDs to human-readable names for notebooks and tags.

This runbook provides comprehensive, step-by-step procedures for managing critical incidents involving Evernote integrations. It covers diagnosing API outages, resolving persistent rate limits, handling authentication failures, and recovering from data synchronization issues. Use this guide during production troubleshooting or service outages to ensure rapid and systematic incident response, including circuit breaker activation and graceful degradation strategies.

A comprehensive guide to boosting Evernote API integration performance and scalability. Techniques covered include implementing response caching (using Redis), performing metadata-only queries to minimize data transfer, batching multiple operations, and optimizing client connection management. Essential for building high-performance, reliable applications.

Provides comprehensive strategies and code examples for managing API rate limits when integrating with the Evernote platform. Techniques include implementing exponential backoff, wrapping clients with minimum delays, batching operations, and optimizing queries (e.g., using metadata only) to ensure robust, high-throughput, and stable API usage.

A comprehensive guide containing minimal working examples for all core Exa search operations. It demonstrates basic metadata search, deep content extraction (text, highlights, summary), finding semantically similar pages, and retrieving content for specified URLs. Ideal for developers starting a new Exa integration or learning advanced RAG patterns.

A comprehensive guide for migrating search pipelines from traditional keyword-based APIs (Google, Bing, Tavily) to Exa's advanced neural search capabilities. This guide covers building an adapter layer, implementing feature-flag traffic shifting, translating legacy query syntax, and leveraging Exa's unique features like semantic and similarity search for superior results.