Skill UI

Empowers building AI agents that perceive screens, plan via vision-language reasoning, and control mouse/keyboard for desktop automation, with attention to sandboxing and safety across Anthropic, OpenAI, and open-source paradigms.

A comprehensive guide to setting up production-grade observability for LangChain applications. It integrates LangSmith tracing for zero-code debugging, custom callbacks for detailed metric collection (latency, token usage, error rate), and Prometheus exporters for robust alerting. This allows developers to monitor LLM pipelines, track performance metrics, and manage operational health in real-time.

Track and improve production LLMs with Langfuse observability: trace API calls, manage prompts, score evaluations, integrate with LangChain/LlamaIndex/OpenAI, and keep dashboards on cost, latency, and quality.

This guide details how to establish comprehensive observability for LLM operations using Langfuse. It covers setting up Prometheus metrics (tracking traces, generation duration, and token usage) and visualizing them in Grafana dashboards. Ideal for monitoring application health, optimizing costs, and tracking performance over time.

TensorBoard delivers a unified interface to track training metrics, debug network behavior, compare experiments, inspect model graphs, and profile performance for PyTorch or TensorFlow workflows.

Assists with tensorboard visualizer tasks for ML training by guiding setup, best practices, and experiment tracking whenever tensorboard visualization is mentioned, producing production-ready configs and validated outputs.

Weights & Biases (W&B) automates experiment logging, real-time dashboards, hyperparameter sweeps, artifact tracking, and model registry/versioning for ML teams using PyTorch, TensorFlow, or HuggingFace workflows.

Hierarchical Agent Memory (HAM) scopes Claude Code agent context by generating per-directory CLAUDE.md files, tracking token savings, and surfacing analytics via a dashboard so multi-folder projects avoid repeated full scans.

A comprehensive protocol defining communication rules and orchestration patterns for multi-agent C-suite teams. It structures complex cross-functional analyses, simulates board meetings, and enforces hard rules—such as depth limits and loop prevention—to ensure consistent, verifiable, and non-circular business insights.

This skill provides comprehensive tools for documenting and managing model performance on Hugging Face. It supports multiple methods for adding structured evaluation data, including extracting tables from README content, importing benchmark scores via APIs (like Artificial Analysis), and executing high-performance custom evaluations using advanced backends such as vLLM or lighteval. It ensures that all metadata is structured and compatible with model leaderboards and industry standards.

A conversational onboarding skill designed to deeply understand the user and define the core identity of an AI partner. Through adaptive, multi-phase dialogue, it extracts key behavioral traits, motivations, and relationship dynamics, culminating in the generation of a comprehensive SOUL.md file that governs the AI's personality and interaction style. Use this skill when setting up, personalizing, or updating your AI agent.

Interactive workflow that interrogates scope, deliverables, constraints, and success criteria while guiding users to refine task prompts; copies the final markdown with Joyride clipboard commands and never writes code.