evaluate-agent-plus-plus

Community

Efficiently evaluate agents with comprehensive analysis and observability

AuthorAesopScott
Version1.0.0
Installs0

System Documentation

What problem does it solve?

This Skill simplifies the evaluation process for AI agents, ensuring thorough analysis, safety checks, and observability, streamlining the handoff for the next phase of development.

Core Features & Use Cases

  • Comprehensive Evaluation: Integrates multiple tools for in-depth agent evaluation, including LangSmith, Inspect AI, and Phoenix.
  • Phase 5 Evaluate Process: Wraps the base Evaluate process for trace-backed eval runs, safety/tool-use checks, and eval reports.
  • Phase 7 Observability Handoff: Provides observability for Phase 7 handoff, aiding in continuous agent improvement.
  • Use Case: Ideal for when Codex needs to create eval suites, trace-backed eval runs, safety checks, or eval reports for equipped agents.

Quick Start

Run the 'evaluate-agent-plus-plus' skill to perform a comprehensive evaluation of an AI agent, including safety checks and Phase 7 handoff preparation.

Dependency Matrix

Required Modules

LangSmithInspect AIPhoenix

Components

scriptsreferencesassets

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: evaluate-agent-plus-plus
Download link: https://github.com/AesopScott/central/archive/main.zip#evaluate-agent-plus-plus

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 620,000+ vetted skills library on demand.