Skill UI

This module provides production-grade patterns for integrating the Juicebox AI-powered people search API. It includes a singleton client for centralized rate-limit handling, a robust request builder for complex criteria (location, company, experience years), and structured error wrappers for reliable API calls (retries, error context). Ideal for building scalable recruitment, market research, or data enrichment tools.

This comprehensive guide assists developers in planning and executing upgrades for systems utilizing the Juicebox API. It covers critical steps such as API version detection, migrating complex data schemas (e.g., search results), updating query syntax, and implementing robust rollback strategies. Essential for maintaining continuous operation in recruiting and market research applications when the underlying API changes.

Loki Mode is a multi-agent autonomous startup system that progresses PRD to production with zero human intervention, orchestrating research-enhanced components such as OpenAI SDK, DeepMind, Anthropic, AWS Bedrock, and agent tooling while enforcing RARV reasoning and efficiency tracking.

Guides researchers through drafting publication-ready machine learning papers for NeurIPS, ICML, ICLR, ACL, AAAI, and COLM by combining expert writing philosophy with LaTeX templates, citation verification APIs, and collaborative workflows.

This comprehensive guide covers the entire MLOps lifecycle, specializing in building robust, scalable, and production-grade Machine Learning systems. Expertise includes orchestrating complex ML pipelines using Kubeflow, Airflow, Prefect, and cloud-native services (AWS SageMaker, Azure ML, GCP Vertex AI). It covers experiment tracking (MLflow, W&B), model versioning, model registries, infrastructure as code (Terraform), and container deployment using Kubernetes. Ideal for engineers tasked with moving ML models from research to reliable, managed production environments.

This skill provides comprehensive monitoring and comparative analysis of cross-chain bridges in the DeFi ecosystem. It tracks key metrics such as Total Value Locked (TVL), transaction volume, typical fees, and real-time transfer status across major networks (e.g., Ethereum, Arbitrum, Polygon). Use this tool to research bridge safety, compare transfer routes before depositing funds, or track specific transaction hashes across multiple protocols like Wormhole, LayerZero, and Stargate.

nanoGPT is a simplified, minimalist implementation designed specifically for educational purposes. It reproduces the core architecture of GPT-2 (124M) using clean, hackable code, allowing users to understand the entire transformer pipeline from scratch. The framework supports full workflows, including data preparation on various datasets (e.g., Shakespeare, OpenWebText), model training, and text generation, making it ideal for students and researchers learning NLP and deep learning principles.

This system implements Role-Based Access Control (RBAC) for OpenEvidence, an AI-powered clinical decision support platform. It strictly manages access to sensitive data, including PHI-adjacent datasets and medical literature, ensuring compliance with regulations like HIPAA. The system defines distinct roles—Admin, Clinician, Researcher, Student, and Auditor—with granular permissions for querying, data export, and institutional configuration, maintaining a secure and auditable data environment.

This skill finds and compares decentralized finance (DeFi) yield opportunities across various protocols. It aggregates crucial data, including APY/APR calculations, Total Value Locked (TVL), and comprehensive risk assessments. Use it when researching yield farming strategies, comparing cross-chain protocols, or optimizing crypto investment returns.

A comprehensive workflow utilizing the Perplexity Sonar API for advanced, multi-turn research. It systematically decomposes broad topics into focused sub-queries, executes batch queries while maintaining context continuity, deduplicates citations, and synthesizes all findings into a structured, comprehensive research document. Ideal for academic research, market analysis, and generating detailed investment or industry briefs.

This guide outlines the comprehensive reference architecture for building advanced, AI-powered search and research applications, modeled after Perplexity. It covers three core tiers: a searchable service with model routing and caching, a citation pipeline for validation and rendering, and an orchestrator for multi-query research synthesis. Use this when designing complex, production-grade, search-augmented LLM integrations or reviewing advanced system patterns.

This skill enables the development of robust, event-driven pipelines interacting with the Perplexity Sonar API. Since Perplexity lacks native webhooks, this system simulates event patterns using advanced techniques: Streaming Server-Sent Events (SSE) for real-time data, job queues (like BullMQ/SQS) for asynchronous batch research, and cron-triggered monitoring for scheduled news and trend alerts. It provides patterns for building complex, reliable, and scalable AI applications.