questionably-ultrathink
Advanced reasoning skill integrating Atom of Thoughts (AoT) and Chain of Verification (CoVe) frameworks. Use when facing complex problems requiring rigorous reasoning, systematic decomposition, or factual verification. Activation triggers: - "be thorough", "analyze carefully", "make sure this is right" - Complex multi-part questions - Architecture or security decisions - "verify", "double-check", "are you sure" - High-stakes technical decisions - Debugging complex issues
npx skills add majiayu000/claude-skill-registry --skill questionably-ultrathink-snowmead-questionably-ultrath-3a1edb34 --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
# UltraThink Reasoning Framework You orchestrate advanced reasoning through isolated, verified atomic solving. <architecture_overview> ## Architecture: Isolated Solving with Question Contraction ### The Problem with Traditional Decomposition Traditional approaches have a critical flaw: the same agent that generates questions also sees all answers. This creates bias contamination - knowledge of other questions/answers influences each response. ### The Solution: True Factored Execution 1. **Graph Generator** creates ONLY the DAG of questions (no solving) 2. **Atomic Solver** answers ONE question per spawn (complete isolation) 3. **Graph Maintainer** rewrites dependent questions with solved answers (contraction) 4. Each solver sees ONLY its contracted question - nothing else ### Flow Diagram ``` SKILL → Graph Generator (DAG only, no solving) ↓ For each level: → Fresh Solver per atom (isolated, only sees question) → Graph Maintainer (contracts dependent questions) ↓ Repeat until FINAL solved ↓ Synthesize response ``` </architecture_overview> <clarification_first> ## Phase 0: Clarify Intent First (MANDATORY) **ALWAYS start by assessing if clarification is needed.** Before invoking any a
- Architecture: Isolated Solving with Question Contraction
- The Problem with Traditional Decomposition
- The Solution: True Factored Execution
- Flow Diagram
- Phase 0: Clarify Intent First (MANDATORY)
- Phase 0.5: Select Analysis Rigor
- Available Agents
- aot-graph-generator
- aot-graph-maintainer
- cove-atomic-solver
- aot-judge (Optional - High-Stakes Only)
- Full Pipeline Orchestration
- Phase 1: Generate Session & Build Graph
- Phase 2: Iterative Solve Loop
What does the questionably-ultrathink skill do?
Advanced reasoning skill integrating Atom of Thoughts (AoT) and Chain of Verification (CoVe) frameworks. Use when facing complex problems requiring rigorous reasoning, systematic decomposition, or factual verification. Activation triggers: - "be thorough", "analyze carefully", "make sure this is right" - Complex multi-part questions - Architecture or security decisions - "verify", "double-check", "are you sure" - High-stakes technical decisions - Debugging complex issues
How do I install it?
Run `npx skills add majiayu000/claude-skill-registry --skill questionably-ultrathink-snowmead-questionably-ultrath-3a1edb34 --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From majiayu000/claude-skill-registry, a repository with 534 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.