complexity-guard

The final gatekeeper. Audits RFCs to reject over-engineering, unnecessary dependencies, and resume-driven development.

Install with Tessl CLI

npx tessl i github:Lingjie-chen/MT5 --skill complexity-guard

What are skills?

1.17x

Quality

62%

Does it follow best practices?

Impact

100%

1.17x

Average score across 3 eval scenarios

Optimize this skill with Tessl

npx tessl skill review --optimize ./.trae/skills/complexity-guard/SKILL.md

SKILL.md

Review

Evals

Evaluation results

100%

10%

Architecture Review: Recipe-Sharing Platform

Architecture over-engineering detection

Criteria

Without context

With context

Report file created

100%

Numeric complexity score

100%

Binary verdict

100%

Microservices Envy flagged

100%

Premature caching flagged

100%

GraphQL fetishism flagged

100%

Monolith-first alternative

100%

REST alternative to GraphQL

100%

Caching deferral alternative

100%

Alternatives are concrete

100%

Without context: $0.3061 · 1m 50s · 11 turns · 130 in / 5,260 out tokens

With context: $0.5697 · 3m 30s · 20 turns · 379 in / 10,033 out tokens

100%

18%

Implementation Plan Review: Inventory Management System

YAGNI and dependency diet enforcement

Criteria

Without context

With context

Report file created

100%

Numeric complexity score

100%

Binary verdict

50%

100%

Plugin system YAGNI flagged

100%

Multi-tenancy YAGNI flagged

100%

lodash flagged

100%

moment.js flagged

100%

Native JS alternative to lodash

70%

100%

Implement current case only

100%

Dependencies individually questioned

100%

Without context: $0.2552 · 1m 42s · 10 turns · 73 in / 4,379 out tokens

With context: $0.4696 · 2m 54s · 19 turns · 170 in / 7,257 out tokens

100%

15%

System Design Review: Notification Service

Generic Hell and Zombie Code detection

Criteria

Without context

With context

Report file created

100%

Numeric complexity score

50%

100%

Binary verdict

100%

Generic base class flagged

100%

Rule of Three violation noted

100%

Zombie Code flagged

100%

Message broker overkill flagged

100%

Concrete class alternative

100%

Simpler messaging alternative

100%

Boring stack endorsed

100%

Without context: $0.2344 · 1m 56s · 8 turns · 57 in / 4,196 out tokens

With context: $0.4226 · 2m 29s · 17 turns · 98 in / 6,764 out tokens

Evaluated: 1 day ago
Agent: Claude Code
Model: Claude Sonnet 4.6

Table of Contents

Architecture Review: Recipe-Sharing Platform Implementation Plan Review: Inventory Management System System Design Review: Notification Service

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.