YCC-Lover/research-starter-skills / skills/research-starter-paper
research-starter-paper
Unverified科研学习与论文写作总入口,协调六个模块。帮助不知道提供什么材料的新手逐步产出、自动保存和恢复项目进度,也用于跨阶段研究任务;区分学习、执行和只读检查。
resources/research-engine/academic-paper/references/writing_judgment_framework.md
View on GitHub# Writing Judgment Framework
A cognitive framework for making writing decisions. Complements the diagnostic `writing_quality_check.md` with judgment heuristics.
## The Clarity Test
For every paragraph, ask: "If I remove this paragraph, does the paper still make sense?"
- **Yes, and nothing is lost** → Delete the paragraph
- **Yes, but context is lost** → Keep, but it's a supporting paragraph — keep it short
- **No, the argument breaks** → This is a load-bearing paragraph — it deserves the most careful writing
**Judgment heuristic**: Most AI-generated academic text fails the clarity test because every paragraph is given equal weight. Vary your investment: load-bearing paragraphs get multiple drafts; supporting paragraphs get one.
## The Reader's Journey
Academic writing is a guided tour. At any point, the reader should be able to answer:
1. **Where am I?** (Section structure, signposting)
2. **Why am I here?** (Connection to the research question)
3. **What should I take away?** (The point of this section)
4. **Where am I going next?** (Transition logic)
If ANY of these is unclear, the writing needs revision — regardless of how accurate the content is.
## Discipline-Specific Voice
| Discipline | Typical Voice | What Makes It Credible |
|-----------|--------------|----------------------|
| Hard sciences | Impersonal, passive, hedged | Precision of measurement and method |
| Social sciences | Semi-personal, active, qualified | Transparency about limitations |
| Humanities | Personal, argumentative, interpretive | Depth of engagement with sources |
| Engineering | Direct, specification-oriented | Reproducibility of results |
| Medical | Structured, protocol-focused | Adherence to reporting guidelines |
**Judgment heuristic**: If you're writing "We argue that..." in a physics paper or "The data indicates..." in a humanities essay, you've adopted the wrong voice for the discipline.
## The "So What" Hierarchy for Sections
| Section | The reader's "So What?" | Your job |
|---------|------------------------|----------|
| Introduction | Why should I care about this topic? | Hook with a real-world consequence or knowledge gap |
| Literature Review | What's missing from what we know? | Build to the gap, don't just summarize |
| Methodology | Can I trust these results? | Show rigor, acknowledge limitations upfront |
| Results | What did you actually find? | Lead with the finding, not the statistical test |
| Discussion | What does this mean for the field? | Connect back to the gap; state the delta clearly |
| Conclusion | What should I remember? | One sentence that captures the contribution |
## Revision Decision Matrix
When revising based on reviewer feedback:
| Reviewer Says | Your Options | Decision Framework |
|--------------|-------------|-------------------|
| "Unclear" | Rewrite for clarity | Always do this — it costs nothing |
| "Missing reference" | Add if relevant | Check if it genuinely fills a gap or is self-citation promotion |
| "Wrong method" | Evaluate, justify, or change | If your method is defensible, justify with evidence; if not, this may require Major changes |
| "Not novel enough" | Strengthen framing | Reframe contribution, don't add unnecessary analyses |
| "Too long" | Cut ruthlessly | Apply the Clarity Test paragraph by paragraph |
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 |