Skills and MCP tagged “llm”

karpathy-guidelines multica-ai skill Guidelines that cut typical LLM coding mistakes: less overcomplication, surgical edits, verifiable outcomes. Claude Code CX Codex Cursor +1 42,500installs google-agents-cli-eval official Google skill Guides testing, dataset setup, and scoring for AI agents via agents-cli. Useful for inspecting failures and running prompt optimization under the Gemini methodology. Claude Code CX Codex Cursor +1 510,000installs caveman-discover JuliusBrussee skill Finds and labels every LLM workflow in your repository so Caveman Cloud breaks down spend by actual job rather than grouping everything into one bucket. Claude Code CX Codex Cursor +1 103,500installs caveman-evidence-review JuliusBrussee skill A read-only skill reviews Caveman Cloud evidence on LLM spend, traces, and routing. Use it to see where a project’s model budget goes. Claude Code CX Codex Cursor +1 103,500installs caveman-setup JuliusBrussee skill Routes LLM API calls through the Caveman Cloud gateway to measure token costs without altering app behavior. Useful for tracking model spend across a repo. Claude Code CX Codex Cursor +1 102,500installs typesafe-ai typesafe-ai skill Guides agents on integrating TypeSafe Decision Models like Jev, answering typed questions about state with calibrated probabilities instead of brittle prompts. Claude Code CX Codex Cursor +1 95,600installs gemini-api-dev official google-gemini skill Guides coding agents to write Gemini API code in Python and TypeScript using current SDKs, interaction endpoints, and active models. Claude Code CX Codex Cursor +1 25,100installs open-code-review alibaba skill Skill runs the ocr CLI to review Git changes with AI. Produces line-level comments and can apply fixes on request. Claude Code CX Codex Cursor +1 6,800installs open-source browser-use skill Reference guide for the browser-use open-source Python library. Helps configure Agent and Browser classes, custom tools, lifecycle hooks, and monitoring. Claude Code CX Codex Cursor +1 3,500installs agent-architecture-audit affaan-m skill This skill checks agent and LLM applications across 12 layers. Use it when an agent degrades and the failing layer is hard to locate. Claude Code CX Codex Cursor +1 4,700installs cost-aware-llm-pipeline affaan-m skill A skill for apps that call LLM APIs: picks a model by task complexity, tracks spend, caches prompts. Useful when your API bill keeps climbing. Claude Code CX Codex Cursor +1 4,600installs prompt-engineering-patterns Seth Hobson skill This skill helps design, test, and refine prompts for LLM applications. Use it for reusable templates, few-shot examples, and structured responses. Claude Code CX Codex Cursor 23,418installs gke-inference official google skill Select a model profile and accelerator, then prepare Kubernetes manifests for AI/ML inference serving on GKE. Claude Code Gemini CLI CX Codex 6,791installs prompt-engineer VoltAgent Community subagent This subagent designs, tests, and improves prompts for LLM production systems. Use it when prompts need evaluation, metrics, or cost tracking. Claude Code 16,078installs prompt-engineer Seth Hobson subagent This subagent designs and tests prompts for LLM features, agent workflows, and system instructions. Use it when prompt structure, safety, or evaluation needs deliberate work. Claude Code CX Codex Cursor 16,078installs deepeval Confident AI skill DeepEval skill sets up evaluation loops for AI agents and RAG systems. It creates pytest suites, generates golden datasets, and catches regressions. Claude Code CX Codex Cursor +1 2,800installs langchain-architecture Seth Hobson skill This skill helps design LangChain 1.x and LangGraph applications with agents, memory, state, and tool calls. Use it for multi-step LLM workflows. Claude Code CX Codex Cursor 13,150installs llm-evaluation Seth Hobson skill A skill for evaluating LLM applications: automated metrics, human feedback, LLM-as-judge and benchmarking. Useful when comparing models or catching regressions before deploy. Claude Code CX Codex Cursor 12,755installs eval-driven-dev official github skill Builds end-to-end evaluations for Python applications that call an LLM, from criteria and test data to scored runs. Use it for QA, benchmarks, and investigating wrong responses. GitHub Copilot 3,900installs ai-engineer Seth Hobson subagent This subagent designs LLM applications, RAG systems, and AI agents. Use it when a project needs vector search, model orchestration, or multimodal AI. Claude Code CX Codex Cursor 7,687installs hindsight-local vectorize-io skill This skill stores preferences, task learnings, and procedure outcomes, then retrieves relevant context before new work begins. Claude Code CX Codex Cursor +1 493installs cli-routing cheaper-inference skill Manage model routing combos from the command line with omniroute. Test routing strategies, inspect metrics, and configure fallback chains interactively. Claude Code CX Codex Cursor +1 1,100installs agents calesthio skill A skill for building ElevenLabs voice agents: assistants, support bots, interactive characters. Fits when you need real spoken dialogue, not just text. Claude Code CX Codex Cursor +1 1,000installs omni-models cheaper-inference skill Queries available AI models across configured providers via OmniRoute. Helps filter models, resolve aliases, and inspect the catalog. Claude Code CX Codex Cursor +1 1,000installs prompt-engineer https://github.com/Jeffallan skill A skill that writes, refactors and tests LLM prompts: templates, JSON schemas, scoring rubrics, test suites. Useful when model output is inconsistent and you need to know why. Claude Code CX Codex Cursor +1 5,600installs omni-cache cheaper-inference skill Skill for viewing LLM cache statistics, clearing entries, configuring TTL policies, and controlling semantic-similarity caching thresholds. Use it to optimize cache performance and monitor cache state. Claude Code CX Codex Cursor +1 961installs hypogenic K-Dense-AI skill Skill to plan and audit HypoGeniC, a tool that generates candidate textual hypotheses from labeled datasets using LLM. Use it if you work with the hypogenic package, its task configs, or HypoBench datasets. Claude Code CX Codex Cursor +1 1,900installs langfuse langfuse skill Equips your coding agent to inspect traces, manage prompts, and run evaluations in Langfuse. Useful when instrumenting LLM calls or building test datasets. Claude Code CX Codex Cursor +1 19,700installs context-window-management sickn33 skill A skill for managing LLM context windows through token counting, history summarization, trimming, and context routing. Claude Code CX Codex Cursor +1 1,500installs arize-prompt-optimization official arize skill This skill analyzes prompts through production traces, evaluations, and annotations. Use it when you need better model responses or want to debug a system prompt. GitHub Copilot 1,100installs arize-instrumentation official arize skill This skill helps add Arize AX tracing to an LLM application. The agent inspects the project first, then proposes instrumentation for approval. GitHub Copilot 1,100installs arize-ai-provider-integration official arize skill Manage Arize AI integrations for language model providers: create connections, inspect their settings, and update or remove them. Use it when configuring models for Arize evaluators or other Arize features. GitHub Copilot 1,100installs ai-engineer sickn33 skill A skill that puts the agent in the role of an LLM engineer, then works from success metrics to RAG pipelines, agent orchestration and the production plumbing around them. Claude Code CX Codex Cursor +1 1,300installs phoenix-evals official [email protected] skill Build and run evaluators for AI and LLM applications with Phoenix. Use it to investigate failures, evaluate RAG systems, and check model output. GitHub Copilot 1,000installs phoenix-cli official arize-ai skill A GitHub Copilot skill that drives the Phoenix CLI to pull traces, spans and sessions, then turns what it finds into a failure taxonomy for your LLM app. GitHub Copilot 1,000installs eval-harness-first Seth Hobson skill Builds the evaluation harness before fine-tuning with goldens, graders, judge calibration, and a base-model baseline. Useful for tracking regressions. Claude Code CX Codex Cursor 2,600installs checkpoint-promotion Seth Hobson skill Gates a trained checkpoint before release by checking data quality, capability drift, forgetting, and paired results against the base model. Claude Code CX Codex Cursor 2,600installs chat-format ruvnet skill Formats chat messages into local model templates like llama3 or chatml for Ollama and llama.cpp, and routes queries across intents using a compact WASM HNSW index. Claude Code CX Codex Cursor +1 742installs agent-tool-builder sickn33 skill A skill for writing tool schemas an agent can actually use: JSON Schema, descriptions, validation, error paths. Useful when it keeps calling the wrong function. Claude Code CX Codex Cursor +1 958installs cost-benchmark ruvnet skill Runs corpus benchmarks to evaluate booster performance and compare spend against Gemini or Claude baselines. Generates verifiable cost tables before releases. Claude Code CX Codex Cursor +1 679installs