eval

Manage eval-driven development workflows for software feature evaluations.

Updated Jan 16, 2026
One-click install
npx skills add https://github.com/Eotel/claude-marketplace --skill eval-eotel
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: eval
Source: https://github.com/Eotel/claude-marketplace/tree/main/plugins/base/skills/eval
Command: npx skills add https://github.com/Eotel/claude-marketplace --skill eval-eotel

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Manage eval-driven development workflows by organizing, executing, and reporting feature evaluations.

Core Features & Use Cases

  • Define eval templates for features that capture capabilities and regression criteria.
  • Run checks against current implementations and log results for traceability.
  • Generate comprehensive eval reports to inform release decisions.

Quick Start

Create a new eval definition for a feature named feature-name and start the define workflow to capture criteria.

Frequently Asked Questions about eval

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
What is eval-driven development and how does it manage feature workflows?

Eval-driven development evaluates software features by defining capability and regression criteria, executing validation checks, and generating auditable logs and reports stored under the .claude/evals directory.

How do I create an eval definition for a new software feature?

To create an eval definition, initiate the define workflow for a feature by capturing its specific capabilities and regression criteria, which the system then saves as a markdown definition file for traceability.

Can I automate regression checks and logging for my current software implementation?

Yes, you can automate regression checks by running eval definitions against your current implementation to validate capabilities, automatically logging the results into markdown files for auditable traceability.

Does the eval workflow require external dependencies or testing frameworks to run?

No external dependencies or testing frameworks are required, as the workflow operates independently by reading and writing markdown definitions and logs directly within the project environment.

How are eval results and logs stored for release decision traceability?

Eval results and logs are stored as markdown files within the project's .claude/evals directory, providing auditable traceability and comprehensive reporting to inform final release decisions.

What is the best way to generate comprehensive eval reports for software capabilities?

The best way to generate eval reports is by executing defined capability and regression checks through the workflow, which compiles the validation logs into comprehensive markdown reports for release decisions.