momo-land/PHD-management-skills / skills/phd-paper
phd-paper
Verified依据真实研究材料建立论文大纲、起草与修订段落、整理返修计划并同步本地论文记录。适用于学术写作和投稿准备;以证据约束引用、实验结果和贡献表述。
| File | Kind | Size |
|---|---|---|
| SKILL.en.md | markdown | 3.1 KB |
| SKILL.md | skill md | 2.7 KB |
| agents/openai.yaml | config | 273 B |
SKILL.en.md
View on GitHub---
name: phd-paper
description: Build outlines, draft and revise academic text, organize revision plans and update local paper records using genuine research material. Use for academic writing and submission preparation while grounding citations, results and contribution claims in evidence.
---
# Paper writing and revision
Read the [local workflow](../phd-manager/references/local-workflow.en.md). Identify the project, paper, target language, manuscript path and requested writing scope from existing context and files. Ask specific questions only when essential material is missing. Reply in English to English requests, while preserving existing research content unless the requested writing or translation task requires changing it.
## Establish the evidence
List `papers` and associated `experiments` and `literature`. Read the specified manuscript and available evidence. Form a concise claim-to-source map: supported conclusions, research hypotheses and unverified references. A title alone does not establish a paper's full findings.
## Perform the requested writing task
- **Outline:** Organize the research question, gap in prior work, method, validation approach, results and limitations. Separate demonstrated contributions from intended ones; record actual section names with `todo` status.
- **Draft:** Follow established terminology and argument structure. Clearly mark missing experimental results rather than filling them with plausible numbers, DOIs, authors or sample sizes. Keep hypotheses labeled as hypotheses and avoid turning correlation into causation.
- **Revise:** Preserve meaning and traceable facts while improving structure, clarity and evidential links. Explain changes to scientific claims. Language editing should not enlarge contributions or minimize limitations.
- **Respond to reviewers:** Distinguish explanations, additional analysis, additional experiments and textual revisions. Record actual change locations and evidence. A response letter must not claim an unperformed experiment has been completed.
- **Prepare submission:** Use supplied or verified venue requirements to organize materials and tasks. `submitted` and `accepted` require evidence of the actual status; a prepared manuscript is not a submitted paper.
## Save and synchronize
1. Edit or create a manuscript version as requested. Without a specified format, save Markdown in `outputs/`. Preserve an existing Word/LaTeX workflow when the required tools are available; disclose missing capabilities rather than pretending a conversion succeeded.
2. Store a concise change summary, manuscript path and evidence gaps in `papers.notes`. Update `status` and genuinely completed `sections` as appropriate, preserving existing sections and associations.
3. Save only the user's own progress percentage. Do not increase it based on generated word count. Unperformed validation remains pending.
4. Save through the CLI and read back. Deliver the manuscript, key changes and evidence gaps. A language-only edit need not force a stage change.
Use [phd-literature](../phd-literature/SKILL.en.md) for source verification and [phd-review](../phd-review/SKILL.en.md) for a separate critical review of the argument.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| add-product-e2e-evalpaperclipai/paperclipAdd or extend a Paperclip full-stack runner E2E workflow, fixture, matcher, or report evidence path for local or Daytona execution. | Stars: 115k | Updated | Claude CodeCodex | Verified | Install |
| add-runner-evalpaperclipai/paperclipAdd or extend a Paperclip Runner protocol evaluation definition, roster, assertion, or report fixture with provenance and narrow validation. | Stars: 115k | Updated | Claude CodeCodex | Verified | Install |
| diagnose-why-work-stoppedpaperclipai/paperclipCompatibility entry point for failed, stalled, looping, or over-recovered Paperclip work. Use fix-failed-paperclip-run for the full diagnosis, regression, repair, real verification, and green-PR workflow; honor an explicitly requested diagnosis-only scope. | Stars: 115k | Updated | Claude CodeCodex | Verified | Install |
| fix-failed-paperclip-runpaperclipai/paperclipDiagnose and fix failed, stuck, slow, or incorrectly blocked Paperclip tasks and runs. Use with a task/run URL or failure report to trace the root cause, prove a regression red then green, check second-order effects, verify the real workflow, and deliver a PR with passing checks. | Stars: 115k | Updated | Claude CodeCodex | Verified | Install |