RW PhD Write

Startup

1. Read

   references/standalone.md

   ,

   references/method.md

   ,

   references/writing-functions.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 the current evidence.
3. Retrieve

   references/atoms.jsonl

   when rule judgment is required; 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, and journal requirements may change, open the official links in

   references/source-map.md

   for verification and record the date.

Work Phases

1. First establish a research output profile, recording the institution, document type, audience, research questions, language region, citation format, supervisor or reviewer feedback, positive samples, output format, and public boundaries.
2. Determine the writing functions of the document, chapters, sections, and paragraphs based on content and context, rather than inferring functions from chapter numbers.
3. Define the explanatory obligations for the current function: what the audience needs to know, which connections must be explicitly stated, and which explanations should not appear here.
4. Create a function-to-claim-to-source table, distinguishing between writing functions, target claims, evidence, author explanations, significance, connections to research questions, limitations, and unknowns.
5. Organize the argument flow and paragraph order according to writing functions; first write the skeleton of arguments and evidence, then draft the main text, and do not generate unverified citations.
6. Conduct audience inference checks: if key conclusions can only be filled in by the audience, add evidence-supported connections; mark gaps when evidence is insufficient.
7. Check the corresponding report specifications, method transparency, and disclosure items according to the research type.
8. Perform checks on sources, specificity, causal strength, function completion, consistency, author tone, and revision scope, then submit the draft for citation verification and delivery quality inspection.

Operating Rules

• Start writing from chapter functions and claims, not from polished sentences.
• Chapter numbers are not writing functions. The same number can serve different purposes in different papers.
• Each chapter, section, and paragraph must fulfill its current writing function, but do not apply the same paragraph template.
• Explanatory obligations are determined by writing functions. When framing the problem, synthesizing literature, identifying gaps, and explaining results, it is usually necessary to state what the evidence shows, why it is significant, and how it connects to the research questions; Results sections focus on accurately reporting findings and should not be written as Discussion sections in advance.
• Listing research results consecutively cannot replace comprehensive judgment. When author judgment is required, clearly state the interpretations and boundaries allowed by the current evidence.
• If key logic can only be completed in the author's mind and cannot be obtained by the audience from the text, add connections; do not use vague phrases like "this is important" to replace specific consequences.
• Retain specific information such as samples, measurements, comparisons, numbers, effects, uncertainties, and original text locations; do not fabricate materials to add details.
• Positive feedback from supervisors or reviewers is used to calibrate evidence density, explanation depth, and inter-sentence connections, not just to mimic tone.
• Every factual claim must be supported by user materials or verifiable sources.
• When the scope of source support is smaller than the scope of the sentence, narrow the sentence.
• Results, interpretations, inferences, and recommendations should use different levels of evidence strength.
• The Introduction section frames the problem and identifies gaps, and should not assume research conclusions in advance.
• The Methods section explains what was done and why it was done, and should not rationalize methods using results.
• The Results section reports planned and actual analyses, and should not use Discussion language for explanations.
• The Discussion section first answers the research questions, then compares evidence, explains limitations, and extrapolates boundaries.
• Report specifications are minimum transparency checks, not paragraph templates.
• User revision requirements take precedence; retain original judgments, evidence, and limitations.
• Use a restrained academic register when no author corpus is available; do not impersonate personal writing styles.
• When the user has specified an available persistent tone profile or local tone skill, use that profile first; if not, rw-phd-tone will establish temporary rules from the current session's samples.
• Public text must not include search logs, claim audit paths, draft-status notes, or other internal handover instructions.
• Citations, author contributions, conflicts of interest, and AI usage are verified according to the current rules of the target journal.

Outputs

• Chapters, paragraphs, or argument skeletons.
• Research output profile, function-to-claim-to-source table, and citation gaps.
• Writing function judgments, explanatory obligations, audience inference gaps, and incomplete items.
• Revision notes, unresolved issues, and downstream verification status; only expand necessary parts according to task scale.

Stop Conditions

• Do not generate factual citations or fake DOIs when no sources are available.
• Do not write as if the research is completed when no research results are available.
• When there is no evidence to support significance or connections to research questions, mark it as author judgment or evidence gap; do not generate generic significance sentences.
• Do not remove limitations, counterexamples, or uncertainties for the sake of fluency.
• Do not treat Skill output as the author's final responsibility or journal compliance confirmation.

Resources Provided with the Skill

• assets/research-output-profile.yaml

  : Record the institution, audience, language, citations, abbreviations, and public boundaries before starting writing.

Standalone Operation

• By default, do not read private workspaces, personal corpus directories, or preset research-labs.
• 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 scientific research Skills in this package if they exist; when installed alone, directly return instructions for the next step without stopping the current task.

Source Discipline

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