cascade-ai-ml-engineering

Behavioral guidelines to reduce common LLM coding mistakes, derived from Andrej Karpathy's observations on LLM coding pitfalls

BioRender MCP server for creating and accessing scientific illustrations

Add AI image generation to Claude Code using Google Gemini's image generation APIs. Generate images from text prompts with watermark removal and background transparency options.

Svelte changes since training cutoff (latest: 5.0.0) — runes ($state, $derived, $effect, $props), snippets replacing slots, callback props replacing events, mount/hydrate API. Load before working with Svelte.

Persistent memory system for Claude Code - seamlessly preserve context across sessions

Agent skills for building agents with LangChain, LangGraph, and Deep Agents

Enables AI agents to use Braintrust for LLM evaluation, logging, and observability. Provides correct API usage, working examples, and helper scripts for common operations.

Recursive Language Model for large context processing with token efficiency

Recursive Language Model integration for Claude Code - intelligent multi-provider routing and unbounded context handling

Persistent diagrams that Claude can read, build on, and export

Skills migrated to ProjectMnemosyne for cross-project reuse

Download and generate plots from Weights & Biases runs (skill + CLI scripts)

Core Hugging Face Hub operations through the hf CLI, including skill installation, repo management, jobs, datasets, models, Spaces, and discovery.

Frontend design skill for UI/UX implementation

NLP Skills Marketplace - LLM fine-tuning 教練式引導工作流程

Skills for working with Langfuse, the open-source LLM engineering platform for tracing, prompt management, and evaluation.

Utility skills layered on top of the core protocols. Browse available skills via /catalog; each skill's SKILL.md carries its own contract.

Front-end expert, proficient in various front-end skills, especially React and Next.js

Hot-reloadable versioned prompts with easy tools for prompt engineering, chain workflows, quality gates. Symbolic syntax: >>prompt --> >>chain @framework :: 'gate'

Opt-in productivity coaching for Claude Code. Use for explicit PUA/try-harder requests, user frustration after repeated failures, requests to retry or change approach, passive/low-quality work complaints, completion checks, evidence requests, test/verification reminders, and Ding-style workplace process cues. Normal calm first-attempt requests are left alone.

State machine-based optimization infrastructure for token reduction through deterministic preprocessing. Features zero-token orchestration, hybrid classification, and specialized LLM agents for prompt optimization and task execution.

Comprehensive .NET backend development with C#, ASP.NET Core, Entity Framework Core, and Dapper for production-grade applications

CLI tool for managing Gemini File Search stores with fully managed RAG

A plugin with smart AI skills

Custom Claude skills for specialized workflows

Local context compression for Claude Code workflows, including structured data, debug-heavy output, and supported source files.