doubt-driven-development

Subject software engineering decisions to adversarial fresh-context review.

4|Updated Jun 10, 2026
One-click install
npx skills add https://github.com/mymx2/foreman --skill doubt-driven-development-mymx2
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: doubt-driven-development
Source: https://github.com/mymx2/foreman/tree/main/.agents/skills/doubt-driven-development
Command: npx skills add https://github.com/mymx2/foreman --skill doubt-driven-development-mymx2

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill prevents costly errors in non-trivial software development by forcing a rigorous, adversarial review of architectural and logic decisions before they are finalized.

Core Features & Use Cases

  • Adversarial Review: Subjects claims and artifacts to a fresh-context, bias-to-disprove examination.
  • Decision Verification: Ensures that non-trivial code changes, architectural shifts, and high-stakes logic are validated against a clear contract.
  • Use Case: When implementing a complex, thread-safe caching layer, use this skill to identify edge cases and hidden coupling that a standard review might overlook.

Quick Start

Apply the doubt-driven-development skill to the current architectural proposal to identify potential failure modes and unstated assumptions.

Frequently Asked Questions about doubt-driven-development

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

FAQPage Schema
What is adversarial code review for software architecture decisions?

Adversarial code review is a fresh-context examination that subjects non-trivial software architecture decisions to a bias-to-disprove analysis, identifying hidden coupling and invalid assumptions to ensure correctness.

How do I verify edge cases in high-stakes logic before implementation?

To verify edge cases in high-stakes logic, apply a five-step doubt cycle: formulate claims, extract artifacts, conduct adversarial questioning, reconcile findings, and iterate within boundaries.

When do I need to use adversarial verification for code changes?

Adversarial verification is needed for non-trivial code changes and complex architectural shifts, such as implementing a thread-safe caching layer, where standard reviews might overlook failure modes.

What is the best way to identify hidden coupling in software architecture?

The best way to identify hidden coupling is using adversarial review that operates on specific artifacts and contracts, applying a fresh-context examination to expose unstated assumptions.

Does adversarial review work without clear contracts and artifacts?

No, adversarial review requires strict adherence to operating on specific artifacts and clear contracts to effectively validate high-stakes development scenarios and identify failure modes.