majiayu000/claude-skill-registry - Page 22

llm-evaluation

Implement comprehensive evaluation strategies for LLM applications using automated metrics, human feedback, and benchmarking. Use when testing LLM performance, measuring AI application quality, or establishing evaluation frameworks.

llm-evaluation

Implement comprehensive evaluation strategies for LLM applications using automated metrics, human feedback, and benchmarking. Use when testing LLM performance, measuring AI application quality, or establishing evaluation frameworks.

llm-function-calling

Implementing function calling (tool use) with LLMs for structured outputs and external integrations.

llm-gateway-routing

LLM gateway and routing configuration using OpenRouter and LiteLLM. Invoke when: - Setting up multi-model access (OpenRouter, LiteLLM) - Configuring model fallbacks and reliability - Implementing cost-based or latency-based routing - A/B testing different…

llmintegration

LLM integration patterns for Claude, GPT, Gemini, and Ollama. Activate for AI API integration, prompt engineering, token management, extended thinking, and multi-model orchestration.

llm-integration

Guide for using LLM utilities in speedy_utils, including memoized OpenAI clients and chat format transformations.

llm-judge-patterns

Comprehensive guide to using LLMs as judges for automated evaluation including prompt patterns, calibration, bias reduction, and multi-judge ensembles

llm-local-deployment

Comprehensive guide for deploying LLMs locally using Ollama, vLLM, and llama.cpp. Local deployment offers privacy, cost control, and reduced latency compared to cloud APIs. This skill covers everythin

llm-ops

Local LLM health checks and cache management. Probe Ollama/vLLM/SGLang endpoints, clean model caches.

llm-pipeline

Pydantic-AI agents, RAG, embeddings for Pulse Radar knowledge extraction.

llm-streaming-response-handler

Build production LLM streaming UIs with Server-Sent Events, real-time token display, cancellation, error recovery. Handles OpenAI/Anthropic/Claude streaming APIs. Use for chatbots, AI assistants, real-time text generation. Activate on "LLM streaming", "SSE",…

llm-streaming

LLM streaming response patterns. Use when implementing real-time token streaming, Server-Sent Events for AI responses, or streaming with tool calls.

llm-token-optimization

See the main LLM Cost Optimization skill for comprehensive coverage of token economics and optimization strategies.

llm-txt-protocol

LLM Text Protocol provides standardized patterns for text-based interactions with language models. It encompasses prompt engineering, response formatting, and text processing patterns that ensure cons

llms-txt-support

Detect and use llms.txt files for LLM-optimized documentation. Use when checking if a site has LLM-ready docs before scraping.

local-llm-fine-tuning

Guides users through the process of preparing datasets and fine-tuning local Large Language Models (LLMs) using techniques like LoRA and QLoRA.

local-rag-search

Local Retrieval-Augmented Generation with vector search.

logfire

Use for Pydantic Logfire observability, tracing, and debugging. Query exceptions, spans, logs with SQL. View traces in Logfire UI. 4 tools for application monitoring and error analysis.

long-document-llm-pipeline

Use when processing documents over 50K characters through LLM APIs with section splitting and batch cost reduction.

longitudinal-drift-detector

Detect behavioral drift and alignment degradation over time.

looking-up-docs

Library documentation via Context7. Use for API references, code examples, framework docs.

lora

Parameter-efficient fine-tuning with Low-Rank Adaptation (LoRA). Use when fine-tuning large language models with limited GPU memory, creating task-specific adapters, or when you need to train multiple specialized models from a single base.

lora-ofl

Imported skill lora_ofl from anthropic

lumen-ai

Master AI-powered natural language data exploration with Lumen AI. Use this skill when building conversational data analysis interfaces, enabling natural language queries to databases, creating custom AI agents for domain-specific analytics, implementing RAG…

lyra

Transform vague inputs into precision-optimized AI prompts for Claude, ChatGPT, Gemini, or other LLMs. Use when user mentions "optimize prompt", "improve prompt", "lyra", "prompt engineering", or needs help crafting effective AI prompts.

mcp-local-rag

Provides score interpretation (< 0.3 good, > 0.5 skip), query optimization, and source naming for query_documents, ingest_file, ingest_data tools. Use this skill when working with RAG, searching documents, ingesting files, saving web content, or handling PDF,…

meta-analyst-agent

AI self-improvement analyst. Tracks AI agent mistakes, analyzes failure patterns, and proposes system improvements. Implements continuous learning loop for trading system enhancement.

Source text: Korean

metacognitive-self-mod

Analyze and improve the improvement process. Use for detecting regressions and meta-optimization.

minimax-mcp-usage

Optimal patterns for MiniMax MCP tools (web_search + understand_image)

moai-ai-nano-banana

Nano-Banana AI service integration for content generation, image creation, and AI-powered workflows. Use when integrating AI services for content creation.

model

Inspect or change Babysitter model-routing policy by phase.

model-ab-test

Use when validating model downgrades for skills or agents. Runs A/B comparison between current and proposed model, scores outputs against test criteria, produces verdict.

model-fallback

模型自动降级与故障切换。当主模型请求失败、超时、达到速率限制或配额耗尽时，自动切换到备用模型，确保服务连续性。支持多供应商、多优先级的智能模型选择，提供健康监控、自动重试和错误恢复机制。

Source text: Chinese

model-management

CLI commands to add, update, list, and deprecate AI models in the database. Triggers on "add model", "new model", "update model", "deprecate model", "model pricing", "model cli".

model-manager

Test, validate, and add new AI models to the eval suite. Use when user asks to add new models, test model access, check pricing, or update models.yml.

model-router

Route work to the cheapest model tier that can reliably complete it.

model-routing

Model selection strategy for subagents — haiku for exploration, sonnet for implementation, opus for architecture/security

model-switch

Switch the agent's model at runtime and restart. Use when the user asks to change models (e.g., "switch to kimi", "use opus", "go cheaper"), or when the agent decides a different model is better for the current workload.

model-usage

Step-by-step guidance for model usage.

model-usage

Use CodexBar CLI local cost usage to summarize per-model usage for Codex or Claude, including the current (most recent) model or a full model breakdown. Trigger when asked for model-level usage/cost data from codexbar, or when you need a scriptable per-model…