← ClaudeAtlas

critical-thinkinglisted

Proactively challenge implementation plans, architecture decisions, and design assumptions. Use when reviewing plans, designs, or technical decisions. Verifies claims via web search, cross-references documentation, identifies risks and gaps, and surfaces hidden assumptions. Activates automatically when evaluating technical proposals. Also the escalation gate — the moment your instinct is to ask the operator, run this skill on the most capable model available and decide whether the question is really theirs.
OneDro1d/dark-factory · ★ 0 · AI & Automation · score 70
Install: claude install-skill OneDro1d/dark-factory
# Critical Thinking Systematically question implementation plans, architecture decisions, and technical assumptions to surface risks, gaps, and unverified claims before they become problems. ## Overview This skill **proactively engages** when reviewing: - Implementation plans and roadmaps - Architecture and design decisions - Technology choices and trade-offs - Performance and scalability claims - Security assumptions - Testing strategies ## Core Principle > **"What evidence supports this decision? What could make it wrong?"** Every technical decision rests on assumptions. This skill surfaces those assumptions and verifies them against: 1. **Web search** — Current best practices, known issues, benchmarks 2. **Existing documentation** — CLAUDE.md, design docs, prior decisions 3. **Architectural principles** — From microservices-architect and any platform-specific service-planner skill 4. **First principles** — Does this make logical sense? ## The VERIFY Framework For every significant claim or decision, apply: | Step | Action | Question | |------|--------|----------| | **V**alidate | Check the source | Where does this claim come from? Is it authoritative? | | **E**vidence | Seek supporting data | What evidence exists? What's the sample size? | | **R**isks | Identify failure modes | What happens if this assumption is wrong? | | **I**mpact | Assess consequences | How severe are the consequences of being wrong? | | **F**alsify | Try to disprove | What would prove this cl