RW Research Question

Activation

1. Read

   references/standalone.md

   ,

   references/method.md

   , and

   references/standards.md

   .
2. Read the materials provided by the user in this session; if no materials are provided, only complete the parts allowed by current evidence.
3. Retrieve

   references/atoms.jsonl

   when rule judgment is needed; read

   references/cases.md

   when encountering similar tasks.
4. Read

   references/domain-guide.md

   when selecting methods or tools, and use

   assets/worksheet.md

   to organize deliverables.
5. Read

   references/acceptance.md

   .

   references/behavior-tests.json

   is only used for testing and not as factual information for user tasks.
6. When current literature, APIs, report specifications, or journal requirements may change, verify by opening the official links in

   references/source-map.md

   and record the date.

Work Phases

1. Retain the original interest, decision-making scenario, and the phenomenon the user actually wants to explain.
2. Identify the problem type: descriptive, correlational, causal, diagnostic, predictive, experiential, implementation, economic, or evidence synthesis.
3. Define the population or context, phenomenon or intervention, comparison, outcome, time, and research level.
4. Write operational definitions, observable indicators, and unacceptable alternative definitions for core constructs.
5. List at least two alternative explanations, one counterexample, and outcomes that would change the conclusion.
6. Output the main question, sub-questions, search concepts, data requirements, and unanswerable parts.

Operational Rules

• Select the problem type first, then the problem framework; PICO is not the default format for all questions.
• For searchable questions, split core constructs into concepts and synonyms.
• For executable questions, specify what data can be obtained, rather than just writing ideal data.
• For falsifiable questions, specify what observations would weaken the main explanation.
• Unclear construct definitions cannot be remedied by sample size or statistical methods.
• If data already exists, only limit the question; do not automatically prove the question is valuable.
• Descriptive questions cannot be turned into causal questions in the output.
• For predictive questions, distinguish between model development, internal validation, external validation, and clinical utility.
• For qualitative questions, specify the context in which experience, process, or meaning occurs.
• For implementation questions, distinguish between intervention effects and implementation outcomes.
• Research questions must have scope boundaries; do not solve problems by adding unlimited variables.
• The final question should be able to generate search strategies, inclusion-exclusion criteria, or entry points for research design.

Output

• Main research question and sub-questions.
• Constructs, boundaries, alternative explanations, and falsification conditions.
• Search concepts, data requirements, and next steps.

Termination Conditions

• Stop at construct clarification when core constructs cannot be defined.
• Mark as non-executable when there is no accessible data or research subjects.
• Do not claim a literature gap based on novel wording.
• Do not rewrite correlational questions into causal conclusions.

Standalone Operation

• By default, do not read private workspaces, personal corpus directories, or preset research-lab.
• Files, text, links, and data provided by the user are inputs for the current task, not installation dependencies.
• When the network is unavailable, continue using stable methods within the Skill; mark parts requiring current facts as pending verification.
• Can connect with other research Skills in this package if available; when installed alone, directly return instructions for next steps without stopping the current task.

Source Discipline

• Separate user materials, public sources, current inferences, and unknown matters.
• Report specifications only check report transparency; do not automatically prove design quality.
• Public source abstracts are stored within the Skill; return to official pages when needing version, cost, policy, API, or journal status.
• Do not generate non-existent papers, data, DOIs, tool operation results, or journal requirements.