skill-eval

Run unit tests and ground truth evaluations for the doc-code-sync skill.

Updated Feb 4, 2026
One-click install
npx skills add https://github.com/stanah/dotagents --skill skill-eval-stanah
Or copy as Structured Prompt for Agentโ–ผ
Please help me install this Agent Skill.
Skill: skill-eval
Source: https://github.com/stanah/dotagents/tree/main/.claude/skills/skill-eval
Command: npx skills add https://github.com/stanah/dotagents --skill skill-eval-stanah

SYSTEM DOCUMENTATION & REQUIREMENTS

๐Ÿ’ก This Skill requires doc-code-sync, python, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill automates the quality evaluation of the 'doc-code-sync' skill by performing a two-layer validation process, thereby saving time on manual evaluations and ensuring consistency in the assessment.

Core Features & Use Cases

  • Layer 1 Validation: Conducts unit tests on individual extractors to ensure functionality.
  • Layer 2 Validation: Evaluates the overall detection rate and accuracy of the skill's workflow against ground truth data.
  • Use Case: When assessing the performance of the 'doc-code-sync' skill, use this Skill to automatically verify the skill's accuracy and completeness of extracted data against expected results.

Quick Start

Use the skill-eval skill to evaluate the 'doc-code-sync' skill's performance with the provided fixtures.

Frequently Asked Questions about skill-eval

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

FAQPage Schema
How do I automate quality assurance validation for doc-code-sync workflows?โ–ผ

Automate quality assurance validation by running a two-layer process that executes unit tests on individual extractors and evaluates the overall workflow detection rate against ground truth data to ensure accuracy.

What is the best way to evaluate accuracy of extracted code documentation against expected results?โ–ผ

Evaluating accuracy requires comprehensive validation against ground truth data, automatically verifying the completeness and accuracy of extracted data to ensure the workflow performs exactly as expected.

How do I run unit tests for data extractors in a Python quality evaluation process?โ–ผ

Run unit tests for data extractors as the first layer of validation, verifying the functionality of individual extraction components to ensure they operate correctly before evaluating the full workflow.

Do I need ground truth data to perform performance testing on doc-code-sync skills?โ–ผ

Yes, ground truth data is a required dependency for performance testing. It serves as the baseline to check the accuracy and completeness of extracted data during the second layer of validation.

Does the skill-eval validation process work without the doc-code-sync dependency?โ–ผ

No, the skill-eval validation process requires the doc-code-sync dependency to function, as it specifically automates the quality evaluation of the doc-code-sync skill's extraction workflow performance.