Paper Draft Generation
Generate a complete IMRaD-structured academic paper draft from research notes and documents.
IMRaD Structure
| Section |
Purpose |
Length |
| Abstract |
Background, objective, methods, findings, conclusions |
250–300 words |
| Introduction |
Problem, motivation, gap, contribution |
400–600 words |
| Related Work |
Prior literature organized by theme |
500–800 words |
| Methods |
How the research was/will be conducted |
400–600 words |
| Results |
What was found (or projected findings) |
300–500 words |
| Discussion |
Interpretation, implications, limitations |
400–600 words |
| Conclusion |
Summary, contributions, future work |
200–300 words |
| References |
All cited sources |
As needed |
Workflow
- Gather content — read all provided notes, wiki pages, and documents
- Identify the core contribution — what is new and why does it matter?
- Map citations — track which sources support which claims
- Draft section by section — follow IMRaD order
- Add inline citations — use
[Author, Year] or [Source Title] format
- Flag gaps — mark sections needing real data with
> ⚠️ [PROJECTED — add empirical data]
Citation format
Use the actual title or author of notes/documents you read:
Predictive coding theory proposes... [Rao & Ballard, 1999]
At the end, add a References section listing all cited sources.
Handling missing data
When the research doesn't have empirical results yet, write projected results based on the hypotheses — but clearly mark them:
> ⚠️ **Projected Results** — Replace with actual experimental data before submission.
Based on the experimental design, we expect...
Quality checklist
1---2name: draft3description: Writes a complete IMRaD-structured academic paper draft (Abstract, Introduction, Related Work, Methods, Results, Discussion, Conclusion, References) from the user's notes, documents, and wiki content. Use when the user asks to write a paper, create a manuscript, generate a draft, or produce academic writing. Do NOT use for short summaries — only for full paper-length output.4license: MIT5---67# Paper Draft Generation89Generate a complete IMRaD-structured academic paper draft from research notes and documents.1011## IMRaD Structure1213| Section | Purpose | Length |14|---------|---------|--------|15| **Abstract** | Background, objective, methods, findings, conclusions | 250–300 words |16| **Introduction** | Problem, motivation, gap, contribution | 400–600 words |17| **Related Work** | Prior literature organized by theme | 500–800 words |18| **Methods** | How the research was/will be conducted | 400–600 words |19| **Results** | What was found (or projected findings) | 300–500 words |20| **Discussion** | Interpretation, implications, limitations | 400–600 words |21| **Conclusion** | Summary, contributions, future work | 200–300 words |22| **References** | All cited sources | As needed |2324## Workflow25261. **Gather content** — read all provided notes, wiki pages, and documents272. **Identify the core contribution** — what is new and why does it matter?283. **Map citations** — track which sources support which claims294. **Draft section by section** — follow IMRaD order305. **Add inline citations** — use `[Author, Year]` or `[Source Title]` format316. **Flag gaps** — mark sections needing real data with `> ⚠️ [PROJECTED — add empirical data]`3233## Citation format3435Use the actual title or author of notes/documents you read:3637```markdown38Predictive coding theory proposes... [Rao & Ballard, 1999]39```4041At the end, add a References section listing all cited sources.4243## Handling missing data4445When the research doesn't have empirical results yet, write projected results based on the hypotheses — but clearly mark them:4647```markdown48> ⚠️ **Projected Results** — Replace with actual experimental data before submission.4950Based on the experimental design, we expect...51```5253## Quality checklist5455- [ ] Abstract covers all 5 components (background, objective, methods, findings, conclusions)56- [ ] Introduction ends with a clear statement of contribution57- [ ] Every major claim has an inline citation58- [ ] Methods section is specific enough to reproduce59- [ ] Discussion addresses alternative interpretations60- [ ] Limitations section is honest, not defensive
richard-kim-79