doubt-driven-development

Automates adversarial fresh-context review of artifacts before finalizing decisions.

6|Updated Feb 17, 2026
One-click install
npx skills add https://github.com/farzanmrz/oparax-chirp --skill doubt-driven-development-farzanmrz
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: doubt-driven-development
Source: https://github.com/farzanmrz/oparax-chirp/tree/main/.claude/skills/doubt-driven-development
Command: npx skills add https://github.com/farzanmrz/oparax-chirp --skill doubt-driven-development-farzanmrz

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Non-trivial decisions are prone to hidden assumptions and overconfidence. This skill enforces a fresh-context adversarial review to surface gaps, test edge cases, and make correctness the default before delivering outputs.

Core Features & Use Cases

  • Fresh-context adversarial review to challenge claims and constraints.
  • Artifact and contract extraction to define what is being evaluated.
  • Adversarial prompt restructuring to force critical scrutiny.
  • Structured reconciliation to track findings and decisions.
  • Safe-decision gating to prevent premature finalization in high-stakes scenarios.

Quick Start

Initiate a doubt-driven review by providing the artifact and its contract to a fresh-context adversarial reviewer and follow the steps before finalizing the decision.

Frequently Asked Questions about doubt-driven-development

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

FAQPage Schema
What is an adversarial code review and when should I use it?

Adversarial review verifies non-trivial decisions by forcing a fresh-context critique to surface hidden assumptions before finalizing outputs. Use it for unfamiliar code, critical systems, and high-stakes changes where correctness matters more than speed.

How do I verify critical code decisions before finalizing high-stakes changes?

Verify critical code decisions by providing the artifact and its contract to a fresh-context adversarial reviewer. This enforces structured reconciliation to track findings and apply safe-decision gating before finalizing outputs.

What inputs do I need to start a doubt-driven development review?

To start a doubt-driven review, you need an artifact, a contract, and an adversarial reviewer prompt. These inputs define what is being evaluated and guide the reconciliation of findings before finalizing the decision.

Does this adversarial review approach work for general software engineering tasks?

This adversarial review approach targets high-stakes coding and critical systems rather than general software engineering tasks. It applies to scenarios where correctness matters more than speed and non-trivial decisions require verification.

When should I avoid using a fresh-context critique for code review?

Avoid fresh-context critique when speed matters more than correctness. This adversarial review process enforces safe-decision gating that intentionally prevents premature finalization, making it unsuitable for rapid, low-risk development.