atomic-thought
INVOKE for research tasks and multi-source understanding. Produces visible dependency graphs and contracted answers. Use during research phase of workflow. Triggers: researching topics, gathering information before building, combining facts from multiple independent areas.
When & Why to Use This Skill
Atomic Thought (AoT) is a sophisticated Claude skill designed for complex research and multi-source information synthesis. It optimizes reasoning by decomposing intricate problems into independent atomic units and visualizing the process through dependency graphs. By systematically solving sub-questions and contracting answers, it ensures high-precision results for multi-hop queries and cross-disciplinary investigations, significantly improving the depth and accuracy of AI-driven research.
Use Cases
- Complex Multi-hop Queries: Efficiently solving research questions that require connecting multiple facts from disparate sources or independent knowledge areas.
- Pre-Implementation Research: Gathering and synthesizing technical, market, or academic data to inform decisions before starting a development or business project.
- Cross-Disciplinary Synthesis: Investigating topics that span multiple fields by breaking them down into manageable, independent sub-problems to ensure factual accuracy.
- Structured Problem Solving: Visualizing the reasoning path through dependency graphs to maintain clarity and prevent context degradation in long-form analysis.
| name | atomic-thought |
|---|---|
| description | "INVOKE for research tasks and multi-source understanding. Produces visible dependency graphs and contracted answers. Use during research phase of workflow. Triggers: researching topics, gathering information before building, combining facts from multiple independent areas." |
Atomic Thought (AoT)
Decomposes problems into independent atomic units that can be reasoned about without historical context dependency.
MUST Invoke When
- Researching a topic before implementation
- Combining facts from 2+ independent knowledge areas
- User asks to "understand", "research", "investigate", or "look into" something
- Gathering information to inform a decision
- Multi-hop queries requiring information from multiple sources
Output Commitment
This skill produces visible structured output:
- Dependency graph of sub-questions
- Explicit answers to each atomic unit
- Contracted final answer with clear provenance
Do NOT use extended thinking alone—invoke this skill to show your decomposition.
Core Mechanism
Apply a two-phase cycle until the problem is directly solvable:
- Decomposition: Break the current question into a dependency graph of sub-questions
- Contraction: Solve independent sub-questions, then contract answers into a simplified atomic question
Process
1. Identify the question's component parts
2. Map dependencies: which sub-questions require answers from others?
3. Solve independent (leaf) sub-questions first—each answer must be self-contained
4. Contract: Reformulate the original question using obtained answers
5. Repeat until directly answerable
Key Principles
- Markov property: Each atomic state contains everything needed to solve it
- No context accumulation: Solved sub-problems contract into the question, not into growing context
- Independence first: Prioritize sub-questions that don't depend on others
When to Apply
- Multi-hop factual queries requiring information from multiple sources
- Research tasks with distinct knowledge areas
- Problems where irrelevant context degrades reasoning quality
- Tasks parallelizable across independent sub-problems
Example Pattern
Original: "What is the GDP difference between the birthplace of the inventor of the telephone and the country where pizza originated?"
Decomposition:
├── Q1: Who invented the telephone? [independent]
├── Q2: Where was [Q1 answer] born? [depends on Q1]
├── Q3: Where did pizza originate? [independent]
├── Q4: What is GDP of [Q2 answer]? [depends on Q2]
└── Q5: What is GDP of [Q3 answer]? [depends on Q3]
Solve Q1, Q3 → Contract → Solve Q2, Q5 → Contract → Compute difference
Anti-Patterns
- Using for simple single-hop questions (overhead not justified)
- Maintaining full reasoning history across atomic units
- Creating dependencies where none exist