skills.rest
rdleclerc
rdleclercCOMMUNITY
Explore all skills by this author →
GitHub

uberskillevolver

Capture evidence-backed lessons from agent runs and promote high-value skill improvements.

Updated Apr 30, 2026
Search Tags:#regression prevention#agentic workflows#evidence-based#evals#skill evolution#validator design#learning records
One-click install
npx skills add https://github.com/rdleclerc/agentic-uber-skills --skill uberskillevolver
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: uberskillevolver
Source: https://github.com/rdleclerc/agentic-uber-skills/tree/main/uberskillevolver
Command: npx skills add https://github.com/rdleclerc/agentic-uber-skills --skill uberskillevolver

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Prevents skills and agent workflows from drifting by capturing post-run lessons with evidence, then promoting only high-value, human-reviewed improvements.

Core Features & Use Cases

  • Evidence-first learning loop: separates observations from lessons and links decisions to concrete run evidence.
  • Promotion gate to avoid bloat: converts only repeated or severe patterns into durable changes (eval seeds, validators, templates, or deletion/simplification).
  • Anti-regression coverage: includes structured checks for scope-fidelity failures, completion-claim regressions, red/green false-green issues, and runtime topology lessons.

Quick Start

Use uberskillevolver after a Tier 2/3 ubergoal run that changed a skill, surprised you, or failed, and produce a post-run learning record that decides what to promote, defer, or delete based on benefit >> cost.

Frequently Asked Questions about uberskillevolver

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

FAQPage Schema
How do I prevent regression in agentic workflows after modifying prompts or playbooks?▼

Promotion gates prevent workflow bloat by converting only repeated or severe patterns into durable changes. They enforce a benefit-cost ratio, ensuring high-value improvements like eval seeds or validators are promoted while low-value changes are deferred or deleted.

What is evidence-based skill evolution for multi-agent coding protocols?▼

You should run skill evolution after a ubergoal-managed execution in Tier 2 or Tier 3 scenarios where a run changed a skill, surprised you, or failed. This timing ensures you capture meaningful deviations while the context is fresh.

How do I structure learning records to avoid false-green test issues in agent workflows?▼

Structure learning records to separate observations from lesson candidates, then apply anti-regression checks for red-green false-green issues and runtime topology lessons. This prevents silent self-modification and ensures only validated changes are promoted.

Does uberskillevolver require a specific environment or prior execution setup?▼

Yes, uberskillevolver requires a prior ubergoal-managed execution in Tier 2 or Tier 3 scenarios. You must provide a learning record that separates observations from lesson candidates to pass through the promotion gate.

Related Skills

Explore More Software Engineering Skills→
outfitter-devoutfitter-dev
28

research-and-report

Automate systematic multi-source research and generate evidence-based recommendations.

Official
Advanced
zkysar1zkysar1
5

curriculum-gates

Evaluate curriculum graduation gates and promote stages when criteria pass.

Community
Advanced
shaharKeisarAppsshaharKeisarApps

compose-expert

Implement cross-platform UI components with Compose Multiplatform for Android, Desktop, and iOS.

Community
Advanced
DuqueOMDuqueOM
5

model-retrain

Automate model retraining workflows with quality gates and MLflow promotion.

Community
Advanced