evaluation

github:craigtkhill/atui-tools · first committed 2025-11-26 · source · badge

Guidelines for creating evaluation suites, including spec.md templates, rubric structures, and code-based validation patterns. Use this when creating new evals or updating existing ones.

Skill Health: not yet scored

Listed, not yet scored. This skill has no public model test yet. It is in the scoring queue; submitted skills go first, then the most-installed.

Version history

4 commits to this file on GitHub (newest first).

2026-02-22 · commit 1edc09c · refactor(skills): change spec format to yaml and extract acceptance-test skill
2025-12-03 · commit 24ca6ff · refactor(skills): update skill descriptions to include specific triggers for when they should be invoked
2025-11-26 · commit c77e78b · docs(evaluation): add rubic.md instructions
2025-11-26 · commit 2deebaf · feat(evaluation): create eval skill

Archived snapshots