OpenBlog
Open-source AI-powered blog generation system — 12-stage pipeline for producing high-quality, AEO-optimized articles.
✨ Features
- 🔄 12-Stage Pipeline — Modular, testable stages from data fetch to HTML output
- 🎯 AEO Optimization — Built-in Answer Engine Optimization scoring (70-85+ scores)
- 🔍 Smart Citations — Automatic source validation and formatting
- 📝 Rich Content — FAQ/PAA extraction, internal links, ToC generation
- 🖼️ Image Generation — AI-powered featured images via OpenRouter
- 🌐 Multi-language — Supports multiple languages with auto-detection
- ⚡ Fast — 60-90 second generation time
📋 Pipeline Stages
Stage 0: Data Fetch & Company Detection
Stage 1: Prompt Construction
Stage 2: AI Content Generation (Gemini + tools)
Stage 3: Structured Data Extraction
Stage 4: Citation Validation ─┐
Stage 5: Internal Links │ (parallel)
Stage 6: Table of Contents │
Stage 7: Metadata │
Stage 8: FAQ/PAA Enhancement │
Stage 9: Image Generation ─┘
Stage 10: Cleanup & Validation
Stage 11: HTML Generation & Storage
🚀 Quick Start
Installation
git clone https://github.com/federicodeponte/openblog.git
cd openblog
pip install -r requirements.txt
Environment Variables
# Required
GEMINI_API_KEY=your_gemini_api_key
# Optional - DataForSEO fallback (when Google Search quota exhausted)
# Cost: $0.50 per 1,000 queries - Standard mode (async task-based)
# Automatically activates when Google Search returns 429/quota errors
DATAFORSEO_LOGIN=your_dataforseo_email
DATAFORSEO_PASSWORD=your_dataforseo_password
# Optional - for Google Drive integration
GOOGLE_SERVICE_ACCOUNT_JSON='{"type":"service_account",...}'
GOOGLE_DELEGATION_SUBJECT=user@domain.com
API Usage
# Start the server
uvicorn service.api:app --reload
# Generate a blog
curl -X POST http://localhost:8000/blog/write \
-H "Content-Type: application/json" \
-d '{
"primary_keyword": "AI in customer service",
"company_url": "https://example.com"
}'
Python Usage
from pipeline.core.workflow_engine import WorkflowEngine
from pipeline.core.execution_context import ExecutionContext
engine = WorkflowEngine()
context = ExecutionContext(
job_id="test-123",
job_config={
"primary_keyword": "AI adoption in customer service",
"company_url": "https://example.com",
},
)
result = await engine.execute(context)
print(result.final_article["Headline"])
🏗️ Project Structure
openblog/
├── pipeline/
│ ├── blog_generation/ # 12 stages (stage_00 to stage_11)
│ ├── core/ # Workflow engine, execution context
│ ├── models/ # Data models, AI clients
│ ├── processors/ # HTML, citations, sitemap
│ ├── prompts/ # Prompt templates
│ └── utils/ # AEO scorer, helpers
├── service/
│ ├── api.py # FastAPI endpoints
│ └── image_generator.py # Image generation
├── tests/ # Test suite
├── docs/ # Documentation
├── modal_deploy.py # Modal deployment
└── requirements.txt
📊 Output Quality
- AEO Score: 70-85+ / 100
- Generation Time: 60-90 seconds
- Citation Validation: Automatic URL checking
- HTML Output: Clean, semantic markup
🔧 Deployment
Modal (Recommended)
pip install modal
modal deploy modal_deploy.py
Docker
docker build -t openblog .
docker run -p 8000:8000 openblog
📖 Documentation
🧪 Testing
pytest tests/ -v
pytest tests/stages/test_stage_00.py -v # Test specific stage
📝 License
MIT License - see LICENSE for details.
🤝 Contributing
Contributions welcome! Please read the contributing guidelines before submitting PRs.
Built with ❤️ by SCAILE
Metadata
Release files for openblog 3.0.1
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| openblog-3.0.1.tar.gz | 494.3 kB | Without provenance | Details for openblog-3.0.1.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| openblog-3.0.1-py3-none-any.whl | Python 3 | none | any | Details for openblog-3.0.1-py3-none-any.whl |
Total release size: 1.0 MB
Release files / openblog-3.0.1.tar.gz
| Download URL | openblog-3.0.1.tar.gz |
|---|---|
| Size | 494.3 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
14282df32c56484bf73a66c5cd724905c10c1a253b1a09f2d4d117f0fa3f0086
|
|
BLAKE2b-256 checksum How to use checksums |
895e8bd7dfed9ad9da76a2b35ccb0d5dc0c4316edb8a278a44d6906fefddb364
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.11.13
|
Release files / openblog-3.0.1-py3-none-any.whl
| Download URL | openblog-3.0.1-py3-none-any.whl |
|---|---|
| Size | 534.8 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
9eeb22fac56906548c619b8fdf3cbd79ae1c820a47816e13de65791483ab989a
|
|
BLAKE2b-256 checksum How to use checksums |
04bd172be65a7349ba8d1ffb2ed1b61c6912a2dd14c706f8c622c223ce8200ec
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.11.13
|
Security for openblog 3.0.1
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.