Generate master-level research papers with real citations in minutes
Project description
OpenDraft Engine
The Python AI engine that powers thesis draft generation. Contains the 19-agent pipeline, citation research, and export functionality.
Structure
engine/
├── draft_generator.py # Main 19-stage pipeline orchestrator
├── config.py # Model settings, API keys, rate limits
├── utils/
│ ├── agent_runner.py # Agent execution engine
│ ├── api_citations/ # Citation APIs (CrossRef, Semantic Scholar)
│ ├── citation_*.py # Citation management & validation
│ ├── export_professional.py # PDF/DOCX export
│ ├── pdf_engines/ # Pandoc, WeasyPrint engines
│ └── deep_research.py # Research phase utilities
├── prompts/
│ ├── 00_WORKFLOW.md # Complete agent workflow
│ ├── 01_research/ # Deep Research, Scout, Scribe, Signal
│ ├── 02_structure/ # Architect, Citation Manager, Formatter
│ ├── 03_compose/ # Crafter, Thread, Narrator
│ ├── 04_validate/ # Skeptic, Verifier, Referee
│ ├── 05_refine/ # Citation Verifier, Voice, Entropy, Polish
│ └── 06_enhance/ # Abstract Generator, Enhancer
└── opendraft/ # CLI tools
Usage
Run Pipeline Directly
cd engine
python draft_generator.py --topic "Your research topic" --level master
Academic Levels
| Level | Words | Chapters | Time |
|---|---|---|---|
| research_paper | 3-5k | 3-4 | 5-10 min |
| bachelor | 10-15k | 5-7 | 8-15 min |
| master | 20-30k | 7-10 | 10-25 min |
| phd | 50-80k | 10-15 | 20-40 min |
Environment Variables
Required in ../.env.local:
GEMINI_API_KEY=your-key # Required
PROXY_LIST=... # Optional: for faster research
SCOUT_PARALLEL_WORKERS=32 # Optional: parallelism
Dependencies
pip install -r requirements.txt
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
opendraft-1.4.1.tar.gz
(147.4 kB
view details)
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
opendraft-1.4.1-py3-none-any.whl
(169.4 kB
view details)
File details
Details for the file opendraft-1.4.1.tar.gz.
File metadata
- Download URL: opendraft-1.4.1.tar.gz
- Upload date:
- Size: 147.4 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.13.2
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
f6719ad33626ba956264bdb383d3f5f766b3dae081a6bc6c4d7d2ddd15d48d58
|
|
| MD5 |
ae4750f734fb719612e965ff5b5581ee
|
|
| BLAKE2b-256 |
707c8060bcffc565838aee9c403c8a60f9eadda7d130d47ecb85e3ba28026a25
|
File details
Details for the file opendraft-1.4.1-py3-none-any.whl.
File metadata
- Download URL: opendraft-1.4.1-py3-none-any.whl
- Upload date:
- Size: 169.4 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.13.2
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
55d9f46144b2296165f76a527d5dfd23d50fb45794a23db988a0016119b805c5
|
|
| MD5 |
a39febc9126c1f7f73a5f94c28ce5edd
|
|
| BLAKE2b-256 |
976a6a5219ac30bd05b6c4826a80b53cd4a01903f4f5dd1f1b18d21627682c35
|