AIMarkTechPostabout 1 hour ago
Anthropic Adds Plugin Evals to Claude Code
4 min read
Anthropic released a plugin evals workflow for Claude Code, adding the `claude plugin eval` command. The tool grades plugin performance against realistic prompts and compares results with a no-plugin baseline to measure skill triggering, resilience to edits, and improvement over a bare model. It runs on Claude Code v2.1.269 or later.
Level
Hype check
Tap to vote and see what everyone thinks.
In this storyAnthropic
