ad-level-up

Validate proposed project rules against anti-overfitting gates and ADRs.

1|Updated May 8, 2026
One-click install
npx skills add https://github.com/alexandremendoncaalvaro/agentic-development --skill ad-level-up
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ad-level-up
Source: https://github.com/alexandremendoncaalvaro/agentic-development/tree/main/.agents/skills/ad-level-up
Command: npx skills add https://github.com/alexandremendoncaalvaro/agentic-development --skill ad-level-up

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill prevents rule-set bloat and overfitting by providing a structured, human-gated workflow for evolving project conventions and engineering standards.

Core Features & Use Cases

  • Anti-Overfitting Gates: Validates every proposed rule against recurrence, generalization, root cause, and cost-effectiveness.
  • Adversarial Review: Uses a multi-lens review process to ensure new rules don't conflict with existing conventions or ADRs.
  • Human-in-the-loop: Ensures no rule is ever written to the repository without explicit, item-by-item user approval.

Quick Start

Invoke the ad-level-up skill to propose a new project rule based on the findings from your latest audit.

Frequently Asked Questions about ad-level-up

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

FAQPage Schema
How do I prevent rule-set bloat and overfitting when adding new engineering standards?

Preventing rule-set bloat requires an adversarial review process that validates proposed engineering standards against recurrence, generalization, root cause, and cost-effectiveness before allowing any writes to the project repository.

What is human-gated rule-set curation for project conventions?

Human-gated rule-set curation is a workflow where proposed conventions undergo multi-lens adversarial review, ensuring no rule is ever written to the repository without explicit, item-by-item user approval to maintain engineering standards.

How do I maintain ADR-0035 machine stores and ADR-0043 project-specific rule layers?

To maintain ADR-0035 machine stores and ADR-0043 rule layers, apply deterministic placement logic and multi-stage adversarial validation to proposed conventions, ensuring rule-set coherence through strict anti-overfitting gate adherence.

Does adversarial review work for evolving existing project rule-sets?

Adversarial review effectively evolves project rule-sets by applying a multi-lens validation process to ensure newly proposed conventions do not conflict with existing ADRs or engineering standards during curation.

Can I automate engineering standards curation without losing human oversight?

You can automate engineering standards curation while retaining human oversight by applying automated adversarial review and anti-overfitting gates, while requiring explicit, item-by-item user approval before any rule is written.

When should I not use automated governance for project rule-sets?

Avoid automated governance for project rule-sets when proposed conventions cannot pass strict anti-overfitting gates evaluating recurrence, generalization, root cause, and cost-effectiveness, or when deterministic placement logic cannot be applied.