Multi-agent research that keeps bringing gifts back — like cats! Dispatch multiple agents to research a topic in parallel, compile findings, and iterate on new discoveries.

|

Meta's 7-8B specialized moderation model for LLM input/output filtering. 6 safety categories - violence/hate, sexual content, weapons, substances, self-harm, criminal planning. 94-95% accuracy....

Named Tmux Manager - Multi-agent orchestration for Claude Code, Codex, and Gemini in tiled tmux panes. Visual dashboards, command palette, context rotation, robot mode API, work assignment, safety...

A hybrid memory system that provides persistent, searchable knowledge management for AI agents (Architecture, Patterns, Decisions).

High-level PyTorch framework with Trainer class, automatic distributed training (DDP/FSDP/DeepSpeed), callbacks system, and minimal boilerplate. Scales from laptop to supercomputer with same code....

Expert guide for understanding the Local Skills MCP server repository - its structure, architecture, and implementation. Use when exploring this MCP server's codebase, understanding how Local...

Merge multiple fine-tuned models using mergekit to combine capabilities without retraining. Use when creating specialized models by blending domain-specific expertise (math + coding + chat),...

Generate specialized skills for each subsystem in the monorepo. Creates shared language skills and subsystem-specific checklists for high-quality AI code generation.

Facebook's library for efficient similarity search and clustering of dense vectors. Supports billions of vectors, GPU acceleration, and various index types (Flat, IVF, HNSW). Use for fast k-NN...

Creates effective SKILL.md files with proper YAML frontmatter, trigger keywords, and best practices. Use when creating skills, writing skill descriptions, authoring SKILL.md files, or learning...

智能笔记与流程图绘制工具（优化版-无权限读取）。根据用户提供的内容，自动生成精美的可视化笔记和流程图，支持多种风格(手写笔记、思维导图、流程图等)，并导出为图片。使用内置模板，无需读取文件权限。适用于：(1) 学习笔记可视化，(2) 知识梳理与总结，(3) 流程图绘制，(4) 概念解释图表

Fast structured generation and serving for LLMs with RadixAttention prefix caching. Use for JSON/regex outputs, constrained decoding, agentic workflows with tool calls, or when you need 5× faster...

Track ML experiments with automatic logging, visualize training in real-time, optimize hyperparameters with sweeps, and manage model registry with W&B - collaborative MLOps platform

Data framework for building LLM applications with RAG. Specializes in document ingestion (300+ connectors), indexing, and querying. Features vector indices, query engines, agents, and multi-modal...

Accelerate LLM inference using speculative decoding, Medusa multiple heads, and lookahead decoding techniques. Use when optimizing inference speed (1.5-3.6× speedup), reducing latency for...

Simultaneous Launch Button - Two-person rule for destructive commands in multi-agent workflows. Risk-tiered classification, command hash binding, 5 execution gates, client-side execution with...

Build autonomous game-playing agents using AI and reinforcement learning. Covers game environments, agent decision-making, strategy development, and performance optimization. Use when creating...

AI 编程系统协调器。触发词："开发"、"实现"、"创建"、"添加"、"修复"、"报错"、"重构"、"优化"、"review"、"评审"、"继续"、"实现"、"为什么"、"记住"、"保存经验"、"复盘"、"分析"、"学习"、"参考"、"模仿"、"/evolve"

Web search and research using Perplexity AI. Use when user says "search", "find", "look up", "ask", "research", or "what's the latest" for generic queries. NOT for library/framework docs (use...