evalview

PackDev tools

Tests your AI agent over time by snapshotting its behavior and flagging changes like altered tool calls or worse output.

Unavailable. Delivery for this kind is on the roadmap — not serving yet.

Add ahel to your AI once: Claude, ChatGPT, Cursor, Claude Code or Codex. Then ask it to use this.

About this app

Behavior regression testing for AI agents. EvalView detects when your agent's behavior drifts e.g changed tool calls, different outputs, or degraded quality even when traditional tests still pass. Snapshot baselines, check for regressions, and auto-heal flaky results, all from inside Claude Code.

Signals

GitHub stars
4k
Forks
312
Last commit
Aug 2026
Advanced
Item type
plugin
Key
anthropics-claude-plugins-community-evalview
Source
github.com/anthropics/claude-plugins-community