Loading...
Loading...
Found 1,650 Skills
Vision, audio, and multimodal LLM integration patterns. Use when processing images, transcribing audio, generating speech, or building multimodal AI pipelines.
Reference guide for permanent free-tier LLM APIs with rate limits, model lists, and OpenAI-compatible integration patterns.
Current LLM prices. How to use the Narev API endpoints — list model pricing (GET) and calculate call cost (POST). Use when the user needs endpoint behavior, parameters, responses, or errors; real-time per-token rates; token-to-USD math for one call; or when they mention "Narev pricing", "model rates", "USD per token", "cost calculation", or "AI unit economics". For committing catalog snapshots or generator scripts, use update-llm-pricing.
This skill should be used when users want to route LLM requests to different AI providers (OpenAI, Grok/xAI, Groq, DeepSeek, OpenRouter) using SwiftOpenAI-CLI. Use this skill when users ask to "use grok", "ask grok", "use groq", "ask deepseek", or any similar request to query a specific LLM provider in agent mode.
Evaluate LLM systems using automated metrics, LLM-as-judge, and benchmarks. Use when testing prompt quality, validating RAG pipelines, measuring safety (hallucinations, bias), or comparing models for production deployment.
Autonomous research review loop using any OpenAI-compatible LLM API. Configure via llm-chat MCP server or environment variables. Trigger with "auto review loop llm" or "llm review".
Use when validating subjective quality criteria that cannot be deterministically tested — applies LLM-based evaluation with structured rubrics for tone, aesthetics, UX feel, documentation quality, and code readability. Triggers: documentation quality check, error message tone review, UX copy evaluation, code readability assessment, design aesthetic review.
Trains large language models (2B-462B parameters) using NVIDIA Megatron-Core with advanced parallelism strategies. Use when training models >1B parameters, need maximum GPU efficiency (47% MFU on H100), or require tensor/pipeline/sequence/context/expert parallelism. Production-ready framework used for Nemotron, LLaMA, DeepSeek.
Karpathy's LLM Wiki: build/query interlinked markdown KB.
LLM-as-judge methodology for comparing code implementations across repositories. Scores implementations on functionality, security, test quality, overengineering, and dead code using weighted rubrics. Used by /beagle:llm-judge command.
Use when setting up, deploying, or operating vLLM Studio (env keys, controller/frontend startup, Docker services, branch workflow, and release checklists).
Maintain a reviewable LLM Wiki from immutable raw notes, including ingest planning, querying, linting, and guarded raw Graphify maps that help agents generate better wiki pages.