tavva/ben-claude-plugins

This skill should be used when the user asks "what evals can we create", "how do I evaluate this", "design an eval", "create evals for", "how do I know if my LLM is working", "measure quality", or mentions evals, evaluation, scoring rubrics, golden datasets, LLM-as-judge, quality metrics, or judge prompts.

This skill should be used when the user asks to "query Langfuse traces", "show sessions", "check LLM costs", "analyse token usage", "view observations", "get scores", "create score", "add score to trace", "query metrics", or mentions Langfuse, traces, or LLM observability. Also triggers on requests to analyse API latency, debug LLM calls, or investigate model performance. Use for prompt management tasks like "list prompts", "get prompt", "create prompt", "update prompt labels", or "deploy prompt to production". Use for dataset management tasks like "list datasets", "create dataset", "add dataset item", "view dataset runs", or "manage evaluation datasets".

This skill should be used when the user asks to "create a README", "write a README", "generate a README", "improve my README", "make my README better", "README best practices", or mentions needing project documentation. Provides guidance for creating excellent READMEs following patterns from awesome-readme.

This skill should be used when the user asks to "create a sprite", "run in sprite", "execute in sprite", "sprite exec", "open sprite console", "list sprites", "destroy sprite", "create checkpoint", "restore checkpoint", "proxy through sprite", or mentions Sprite, isolated environments, or persistent microVMs. Also triggers on requests to manage sprite authentication, checkpoints, or port forwarding.