Preserve the full anti-skimming form and distinguish it from the brief end message. Describe batched changes with distinct predicted effects and conditional Bayesian updates. Replace normalized-score reliability verdicts with raw paired comparisons and explicit uncertainty.
Validation: diff --check; fresh-agent design scenarios exercised batching, normalization and full-form requirements. No actual training run or measured uplift test. Scenario agent placed the full form in the linked run log, leaving chat-display interpretation unresolved.
Co-Authored-By: Pi <288921227+claudypoo@users.noreply.github.com>
- refs/ -> references/, the folder name the Agent Skills spec uses and the one
Hermes skips when it walks for nested skills.
- rl and pinn declared name: rl and name: pinn, which are global names in a flat
skill namespace. Now ml-debug-rl and ml-debug-pinn. They also called themselves
sub-skills of 'ml-debugging', which is not this skill's name.
- Drop the dead link to SKILL_old.md. It moved into gitignored slop/, so the link
was broken for anyone who cloned.
- Route references/llm_judge_litreview.md, the one reference SKILL.md never named.
- Description leads with the trigger situations. Hermes truncates it to 57 chars.
Co-Authored-By: Claudypoo <288921227+claudypoo@users.noreply.github.com>
