Skill UI

A comprehensive framework for rigorously testing and benchmarking LLM agents. It moves beyond simple pass/fail checks to assess complex behaviors, reliability metrics, and capability consistency across multiple runs. Ideal for production monitoring and identifying subtle failure modes in advanced AI agents.

A systematic workflow designed to significantly improve the reliability and performance of existing AI agents. It covers detailed performance analysis, including gathering metrics (e.g., task success rate, latency), classifying failure modes, and advanced prompt engineering techniques (like Chain-of-Thought and Constitutional AI). The process culminates in structured A/B testing and controlled deployment, ensuring measurable, robust improvements.

A comprehensive Python SDK for accessing Azure AI Language service capabilities. It enables advanced Natural Language Processing (NLP) features, including sentiment analysis, named entity recognition (NER), key phrase extraction, language detection, and sensitive PII detection. Ideal for building sophisticated text analysis applications and data pipelines.

A comprehensive Python SDK for accessing Azure AI Language service capabilities. It enables advanced Natural Language Processing (NLP) features, including sentiment analysis, named entity recognition (NER), key phrase extraction, language detection, and sensitive PII detection. Ideal for building sophisticated text analysis applications and data pipelines.

Provides expert guidance on designing, implementing, and optimizing cutting-edge computer vision pipelines. Expertise spans real-time object detection (YOLO26), advanced promptable segmentation (SAM 3), and leveraging Vision Language Models (VLMs) for complex visual reasoning. Ideal for building high-performance systems in robotics, autonomous vehicles, and industrial inspection, including 3D reconstruction and edge deployment optimization.

A comprehensive toolkit for building production-grade AI agents focused on crypto business development. This skill guides users through the entire token listing lifecycle: autonomously discovering promising assets across multiple chains, applying a weighted 100-point scoring model, verifying safety via advanced wallet forensics, and managing outreach pipelines with necessary human oversight and LLM cascade patterns.

Formal evaluation framework for Claude Code sessions that applies evaluation-driven development to define capability and regression checks, run deterministic and model graders, and track pass@k/pass^k reliability metrics before shipping.

Evaluates large language models across 60+ academic benchmarks (MMLU, HumanEval, GSM8K, TruthfulQA, HellaSwag) with standardized prompts and metrics, making it easy to benchmark HuggingFace or API models, compare releases, and track training progress.

Automate Gong conversation intelligence by using Composio MCP natural-language commands to fetch call recordings, transcripts, analytics, speaker stats, and workspace metadata across date ranges.

This advanced skill is a multi-agent research framework designed for executing deep, extensive research in parallel. It orchestrates a team of 10 specialized agents to gather, synthesize, and analyze information simultaneously. It is ideal for complex topics that require rapid, broad, and battle-tested investigation across multiple dimensions, significantly accelerating the knowledge discovery process.

This comprehensive guide details the end-to-end process of building Retrieval-Augmented Generation (RAG) pipelines using LangChain. It covers critical steps: loading diverse document types (PDF, CSV, TXT), splitting text into manageable chunks, generating high-quality embeddings (OpenAI), storing them in various vector databases (FAISS, Pinecone), and finally constructing the full runnable chain for accurate and context-aware querying.

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.