Skill UI

LlamaIndex is a data-centric framework for building LLM applications, offering 300+ connectors, vector indices, query engines, agents, and multimodal support to power document Q&A, chatbots, knowledge retrieval, and RAG pipelines over private data.

LLaVA is an open-source vision-language assistant pairing CLIP vision encoders with Vicuna/LLaMA models to power multimodal chatbots, multi-turn image Q&A, captioning, and instruction-following tasks.

Provides production-ready guidance for constructing retrieval-augmented generation pipelines, agent architectures, and LLMOps best practices, so teams can reliably deploy context-aware AI assistants with vector search, embeddings, and tool use.

This skill guides users through the entire lifecycle of developing intelligent AI assistants, chatbots, and conversational interfaces. It specializes in establishing robust natural language understanding, managing multi-turn context, and defining best practices for creating seamless, production-ready AI applications with real-world value.

This skill guides users through advanced prompt optimization techniques for Large Language Models (LLMs). It functions as an expert prompt engineer coach, applying methodologies like Constitutional AI and Chain-of-Thought reasoning. Use it to transform basic, raw instructions into highly effective, production-ready prompts, significantly improving accuracy, reducing hallucinations, and optimizing token usage for cost efficiency.

Master comprehensive strategies for evaluating Large Language Model (LLM) applications. This skill covers automated metrics (BLEU, ROUGE, BERTScore), structured human evaluation dimensions, and advanced LLM-as-Judge techniques. Use it to systematically measure model performance, compare prompts, detect regressions, and build confidence in production systems.

Practical guide to integrating Claude models: covers API client usage, streaming responses, tool calls, RAG pipelines, document chunking, cost-aware model selection, and anti-pattern warnings for efficient LLM workflows.

Loki Mode is a multi-agent autonomous startup system that progresses PRD to production with zero human intervention, orchestrating research-enhanced components such as OpenAI SDK, DeepMind, Anthropic, AWS Bedrock, and agent tooling while enforcing RARV reasoning and efficiency tracking.

Extends transformer context windows using RoPE, YaRN, ALiBi, and position interpolation so long documents (32k-128k+ tokens) can be processed, pre-trained models lengthened, and efficient positional encodings or extrapolation strategies deployed.

Automates Mailcoach operations through Composio’s Rube MCP toolkit by first discovering current tool schemas, managing the mailcoach connection, and executing schema-compliant workflows safely with session and memory handling.

Guidelines for automating Mailsoftly tasks through Composio’s Mailsoftly toolkit via Rube MCP, covering discovery, connection checks, execution steps, and schema-aware tool usage for reliable workflows.

Mamba is a selective state-space model architecture delivering O(n) inference for million-token sequences, offering hardware-aware Mamba-1 and Mamba-2 variants that beat Transformers on speed without KV cache overhead.