Back to stack
Dec 24, 2025
2 min read

TasqLeveler – The Instruction Enhancer

What if your AI could make itself smarter before it even starts coding? That's exactly what TasqLeveler does.

TasqLeveler

The TasqLeveler agent acts as an “Instruction Enhancer.” Its primary role is to read your project spec (tasq.md), analyze what you’re building, and automatically add the architectural guardrails that turn a 65% success rate into 95%+.

How it works

Before your first cycle even kicks off, TasqLeveler enhances your tasq.md with:

EnhancementWhat It DoesWhy It Matters
📦 Dependency GraphShows what can import whatNo more circular import nightmares
🎯 Golden Path TestsCode that MUST work for each moduleSuccess criteria, not vibes
🧪 Mock InfrastructureMock servers for external servicesTest without real APIs
📋 Success CriteriaGlobal definition of “done”Clear pass/fail, no ambiguity
⏱️ Priority GuideWhat to build if tokens run lowSmart budget allocation
🔗 Base ClassesAbstract bases for similar modulesConsistent architecture

The Numbers Don’t Lie

MetricWithout TasqLevelerWith TasqLeveler
Imports resolve85%95%+
Classes instantiate80%95%+
Tests pass60%80%+
Fully functional65%75-80%

That’s a +15-20% improvement just from better instructions. The AI doesn’t get smarter - it gets better directions.

What’s configurable

config.yaml (under agents.tasqleveler):

  • provider (string): AI provider (e.g., ‘openai’, ‘anthropic’, ‘gemini’, ‘deepseek’, ‘qwen’)
  • model (string): AI model (e.g., ‘gpt-4.1-mini’)