Pıer
TidesCurrentsHarbor LightsLabBottlesAshore
Pıer

Navigation

  • Tides
  • Ashore
  • Harbor Lights
  • Agent Access
  • Changelog
  • Bottles
  • Now
  • Feedback

External links

GitHubCloudborne ↗

© 2026 Pier.

Read original
Hacker News·edonadei·Sep 11, 2026, 10:20 PM

Anthropic Releases CLI for Evaluating Claude Code Plugins and Skills

Original title:Anthropic released a CLI to evaluate skills and plugins

Products74

Anthropic has introduced a dedicated CLI evaluation framework within its Claude Code documentation to test and benchmark custom skills and plugins. The tool allows developers to define test suites that systematically assess tool-call precision, context ingestion, and error recovery in agentic workflows. As modular coding agents expand their capabilities, the release helps transition plugin authoring from ad-hoc prompting into a measurable software engineering process.

Why it's worth reading

As Claude Code ecosystem adopts modular extensions, tool reliability is crucial; this official CLI establishes a standardized testing benchmark for developers before deploying custom agent skills.

Tags

AnthropicClaude CodeCLILLM EvalsAgentDeveloper Tools

Score breakdown

  • Novelty70
  • Impact72
  • Practicality85
  • Credibility90
  • Timeliness80