eval

Generate YAML evaluation specs with building and verification sections.

1|Updated Jan 14, 2026
One-click install
npx skills add https://github.com/harivansh-afk/eval-skill --skill eval-harivansh-afk
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: eval
Source: https://github.com/harivansh-afk/eval-skill/tree/main/skills/eval
Command: npx skills add https://github.com/harivansh-afk/eval-skill --skill eval-harivansh-afk

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps teams generate structured evaluation specs that define both the building work and the verification checks before implementation, ensuring alignment and measurable outcomes.

Core Features & Use Cases

  • Generate .claude/evals/<name>.yaml configurations containing building_spec and verification_spec to guide builders and verifiers.
  • Provide deterministic verification paths and agent-based checks to validate specs ahead of coding.
  • Use cases include defining acceptance criteria, feature scoping, and planning work before implementation.

Quick Start

Create an eval for a feature and run the build and verify loop.

Frequently Asked Questions about eval

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

FAQPage Schema
How do I generate evaluation specs for feature verification?

To generate evaluation specs, you define both the building work and the verification checks in a YAML document. This creates measurable acceptance criteria that align teams on the scope of work before implementation begins.

What is a building_spec and verification_spec in YAML?

A building_spec and verification_spec are YAML configurations that define what to build and how to verify it. They provide deterministic paths and agent-based checks to validate feature specs ahead of coding.

How do I set acceptance criteria for AI feature planning?

You set acceptance criteria by creating an eval configuration that outlines the building and verification requirements. This ensures teams have structured, measurable outcomes aligned before development starts.

Can I use YAML specs to align teams on feature scope?

Yes, you can use YAML specs to align teams on feature scope. The generated evaluation document defines the exact building and verification parameters, ensuring everyone understands the work and checks required.

Does the eval skill require external dependencies to run?

No, the eval skill requires no external dependencies to run. It independently generates the YAML configurations needed to guide builders and verifiers through the feature planning and checking process.

What is the best way to define deterministic verification paths for features?

The best way to define deterministic verification paths is by generating a structured evaluation spec. This YAML document pairs the building instructions with specific verification checks to validate features ahead of coding.