AI Engineering

A structured process for feature development that prioritizes understanding the codebase, asking clarifying questions, comparing architecture trade-offs, and conducting multi-agent code reviews.

A guide for creating effective skills that extend Claude's capabilities with specialized knowledge, workflows, or tool integrations. Use when building or updating custom skills.

Provides REST API design best practices and auto-generates endpoints with framework-specific routing, validation, and error handling.

Procedural memory system for AI coding agents with confidence decay, anti-pattern learning, and cross-agent knowledge transfer. Use before tasks to retrieve relevant rules, anti-patterns, and history.

Autonomous memory for AI agents that observes conversations, compresses them into prioritized notes, and consolidates older entries. Recovers missed sessions and injects context at startup without manual saving or databases.

Adopt the cognitive framework and expression style of Justin Sun, based on analysis of over 21,000 tweets, autobiography, audio courses, and interviews. Use for crypto industry analysis, business strategy, and decision-making feedback through 14 mental models and 19 heuristics.

Applies Naval Ravikant's mental models to provide advice on wealth, leverage, and specific knowledge, emulating his oracle-like speech patterns.

Automates Chinese contract review with a 7-step workflow, six-dimensional risk evaluation, and generates review documents. Suitable for mainland China civil and commercial contracts.

Analyzes legal issues through the lens of Chinese law, applying statutory interpretation, guiding case reference, and claim basis analysis to understand rights and obligations, assess liability risks, identify applicable legal provisions, and recommend compliance strategies.

A network for AI agents to discover each other, communicate securely, and transact on-chain.

Spawn parallel AI workers for any long-running batch API task. Auto-calculates optimal worker count and stagger, with tiered delegation to reduce costs.

Practice communication with a simulated crush persona built from your chat history, with real-time or post-session coaching feedback.