Skill UI

A comprehensive evaluation suite designed to benchmark the performance of Large Language Models (LLM) and Vision-Language Models (VLM) specifically for home security applications. It covers 143 diverse tests across 16 critical domains, including tool use, prompt injection resistance, contextual reasoning, alert triage, and detailed scene analysis, ensuring robustness for smart home AI systems.

This tool enforces four rigorous coding principles, inspired by Andrej Karpathy's insights on LLM pitfalls. It guides developers to surface assumptions, prioritize simplicity, make surgical changes, and define verifiable goals. Use it when reviewing code, checking diffs, or before committing to ensure high code quality and prevent over-engineering.

A deep dive into handling typed content blocks (text, image, tool_use, thinking, document) in LangChain 1.0. This skill standardizes multi-modal message composition, tool-call iteration, and complex content handling across providers like Claude, GPT-4o, and Gemini, solving key API discrepancies.

This guide provides advanced patterns for composing robust, multi-step Large Language Model (LLM) workflows using LangChain 1.0 core runnables. Learn to implement parallel retrievals (RunnableParallel), conditional routing (RunnableBranch), and state management. It also addresses common pitfalls like silent KeyErrors due to mismatched dictionary shapes, providing solutions using debug probes and type annotations for reliable RAG, classification, and extraction pipelines.

This harness provides comprehensive, reproducible evaluation pipelines for complex LLM chains and agents (LangChain/LangGraph 1.0). It integrates golden dataset management, LangSmith evaluation, RAGAS metrics, deepeval LLM-as-judge, and structured agent trajectory analysis. Use it when establishing quality benchmarks for new chains, diagnosing performance regressions after model switches, or implementing CI/CD gates to prevent quality drops.

This skill provides a comprehensive guide to building robust LangGraph 1.0 StateGraphs in Python. It covers defining typed TypedDict states with reducers, writing deterministic nodes, and defensively wiring conditional edges. It addresses common pitfalls such as silent graph termination, incorrect state updates using reducers, proper management of recursion limits, and handling evolving checkpoint schemas for reliable LLM workflow orchestration.

This skill defines canonical, ordered middleware patterns for LangChain and LangGraph agents. It addresses critical production-level LLM deployment issues, including PII leakage across tenants, token budget overconsumption, prompt injection defense (Guardrails), and cache key corruption. Use this to build secure, scalable, and cost-aware AI applications.

A comprehensive guide to advanced, production-ready LLM inference using LangChain 1.0. This skill addresses common pitfalls and bugs when integrating multiple providers (Claude, GPT-4o, Gemini), covering issues like structured output reliability, correct streaming token accounting, and robust handling of complex AIMessage content blocks. It teaches best practices for model initialization, provider routing, and version-safe code implementation.

This skill integrates LangChain 1.0 and LangGraph 1.0 into standard OpenTelemetry (OTEL) backends (e.g., Jaeger, Honeycomb, Tempo). It solves critical observability gaps by ensuring correct prompt content capture, propagating spans across complex subgraphs, and providing five LLM-specific Service Level Objectives (SLOs) like latency and cost. Ideal for multi-tenant, compliance-sensitive, or existing OTEL stack environments where LangSmith is unsuitable.

This skill provides advanced techniques for robust prompt engineering within LangChain applications. It solves common production issues, such as f-string template breakage when user input contains literal curly braces, and guides developers on consolidating scattered prompts. Crucially, it details how to use the LangSmith Prompt Hub to version, pin, and A/B test prompts using immutable commit hashes, ensuring stable, reproducible LLM deployments.

This skill transforms an LLM into a disciplined wiki maintainer for Obsidian. It moves beyond simple RAG by incrementally ingesting sources (papers, notes, articles) to build a persistent, interlinked "Second Brain." The LLM manages cross-references, updates core concepts, and maintains an evolving knowledge graph, ensuring that knowledge compounds over time rather than being re-derived on every query. Ideal for deep research, personal knowledge management (PKM), and building comprehensive internal wikis.

This bundle collects comprehensive diagnostic evidence from the MindTickle sales enablement API. It gathers data on API token validation, course catalog status, user enrollment, assessment progress, and CRM sync state. It allows users to diagnose complex issues like content gaps, scoring discrepancies, and SSO failures without requiring admin access, providing a complete diagnostic tarball for support teams.