NeVibeCode · 50 items
Skills and MCP tagged “llm”
prompt-caching skill Skill for optimizing LLM work through prompt caching, response caching, and Cache Augmented Generation. Reduces latency and API cost on repeated queries with identical context.
Claude Code CX Codex Cursor +1
867installs
agents skill The skill builds voice AI agents on ElevenLabs: sets first_message, prompt, LLM and voice through CLI or SDK. Useful for voice assistants and support bots.
Claude Code CX Codex Cursor +1
11,000installs
prompt-optimizer
official skill Converts rough drafts, tasks, and half-formed ideas into finished, copy-pasteable chat prompts without placeholder variables.
GitHub Copilot
467installs
llm-architect subagent Designs production LLM architectures: covers RAG, LoRA fine-tuning, inference serving with vLLM, and multi-model routing. Essential when managing token budgets and latency targets.
Claude Code
649installs
senior-ml-engineer skill Guides production ML deployment, MLOps pipelines, RAG systems and LLM API integration. Use it when a project needs monitoring, rollout checks or request cost controls.
Claude Code
915installs
llm-cost-optimizer skill Audits token usage and cuts LLM API spend through model routing, caching, and prompt compression. Useful when API bills spike or before launching new AI endpoints.
Claude Code
73installs
ai-sdk
official skill Provides Vercel AI SDK guidance for your coding agent. Helps build chat interfaces, streaming routes, tool calls, and agents without obsolete syntax.
CX Codex GP ChatGPT
22installs
ai-generation-persistence
official skill Shows how to store LLM outputs with IDs, metadata, and addressable URLs. Useful when an app needs generation or chat history that survives a page refresh.
CX Codex GP ChatGPT
5installs
phoenix
official plugin Add AI observability, evaluation, and OpenInference tracing tools to GitHub Copilot. It helps when you need to inspect traces or debug LLM apps.
GitHub Copilot
★ 40,369repo stars
arize-ax
official plugin The plugin adds Arize AX skills for tracing, evaluating, and tuning LLM applications. Use it to inspect traces, run experiments, and optimize prompts.
GitHub Copilot
★ 40,369repo stars
41–50 of 50