skills.rest
azishio
azishioCOMMUNITY
Explore all skills by this author →
GitHub

anti-overengineering-gain

Displays upstream benchmark results for the anti-overengineering ruleset as a plain-text scoreboard.

Updated Jul 11, 2026
Search Tags:#benchmark#anti-overengineering#scoreboard#coding agent#code reduction#token cost
One-click install
npx skills add https://github.com/azishio/coding-skills --skill anti-overengineering-gain-azishio
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: anti-overengineering-gain
Source: https://github.com/azishio/coding-skills/tree/main/skills/anti-overengineering-gain
Command: npx skills add https://github.com/azishio/coding-skills --skill anti-overengineering-gain-azishio

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Users evaluating the anti-overengineering ruleset lack a quick way to see its measured impact on code size, token usage, cost, and speed without reading the upstream benchmark reports. ## Core Features & Use Cases - Benchmark Scoreboard: Renders a compact plain-text summary of the upstream ponytail v4.9.0 benchmark results, covering lines of code, tokens, cost, time, and safety. - Honesty Boundary: Explicitly labels figures as upstream measurements and refuses to invent per-repo savings numbers, pointing to /anti-overengineering-debt for real counted data. - Use Case: When deciding whether to adopt the anti-overengineering skill family, invoke this skill to see that the upstream agentic benchmark showed 54% fewer lines of code, 22% fewer tokens, and 20% lower cost with no safety trade-offs. ## Quick Start Ask the agent to show the anti-overengineering gain scoreboard to view the upstream benchmark impact summary.

Frequently Asked Questions about anti-overengineering-gain

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

FAQPage Schema
How much code does the anti-overengineering ruleset save?▼

The upstream ponytail benchmark measured 54% fewer lines of code on average in agentic runs, and 80-94% in single-shot tasks. These are upstream figures, not measurements from this repository.

How do I see the anti-overengineering impact scoreboard?▼

Invoke the skill with /anti-overengineering-gain or ask what anti-overengineering saves. It renders a one-shot plain-text scoreboard covering code, tokens, cost, time, and safety, then changes nothing.

Are the anti-overengineering gain numbers measured in my repository?▼

No, the figures come from the upstream ponytail v4.9.0 benchmark of its full intensity, not this repo or this exact skill. For real per-repo numbers, use /anti-overengineering-debt, which counts deferred shortcuts in a ledger.

Does reducing code with anti-overengineering cut safety checks?▼

No, the upstream benchmark reported 100% safety retention: no validation, error handling, security, or accessibility was cut. The ruleset explicitly forbids simplifying away trust-boundary validation or data-loss prevention.

Does the anti-overengineering gain skill change any files or settings?▼

No, it is a one-shot display that edits nothing and persists nothing. It only renders the scoreboard and points to the debt and review skills for follow-up analysis.

Related Skills

Explore More Software Engineering Skills→
NzettodessNzettodess

ponytail-gain

Displays a one-shot ASCII scoreboard of ponytail's benchmarked code, cost, and speed reductions.

Community
Basic
amanpal3amanpal3
1

ponytail-gain

Displays a benchmark scoreboard comparing code size, cost, and speed with and without ponytail.

Community
Basic
avbelavbel
2

dev-code-style

Guides writing compact, readable code with justified abstractions and purposeful comments.

Community
Intermediate
TyrizxTyrizx

update-screenshots

Updates committed blocks-ci screenshot hashes from CI job output after check failures.

Official
Intermediate