BIRDY-EDWARDS 2.0
Infiltrate & Expose
Automated AI powered Facebook intelligence tool for target profiling, network analysis and threat reporting. Runs entirely on-device via Ollama. AI-powered Facebook SOCMINT platform — 100% local, zero cloud dependency. For lite version check this repo click here
Documentation
Full usage guides for both investigation modes are available on the project website:
How it works
What's new in v2.0
- 🗽 Flexibility - Now you can stop investigation in any phase and generate report pdf
- 🎬 Reel intelligence — AI extracts context summary, topic, named entities, and hashtags from every reel caption
- 🔗 Co-commentor coordination matrix — Heatmap of who comments alongside whom. Click any cell to see shared posts
- 👤 HOG + CNN hybrid face detection — Auto model selection by image size with CNN→HOG fallback. Face detection now runs on text post screenshots too
- 📬 Telegram notifications — PDF report automatically sent to your configured Telegram chat on completion
- 📄 Full PDF suite — Added image intelligence, reel intelligence, face gallery, post timeline, co-commentor pairs sections. Bengali, Hindi, Arabic, Urdu render correctly
- ⚙️ Investigation queue — Up to 5 investigations queued and auto-executed sequentially
- 🔄 Real-time progress page — Live stat counters unlock as data arrives during the pipeline
- 📅 Date and Time Modification - During scraping it may possible that we gather the date only not exact timestamp, so as an investigator if you want to modify this into exact timestamp it is possible now.
Features
- 🔍 Profile collection — Automated data gathering of posts, photos, reels, about data, comments, commentor profile links and names with 9-strategy date extraction
- 🗂️ Batch investigation — Submit up to 10 mixed post URLs and run full AI pipeline across all of them
- 🧠 Interaction intelligence — AI sentiment, stance, emotion, and language analysis per interaction in 140+ languages including Bengali, Hindi, Arabic, Urdu, Kuki
- 📊 Actor scoring — Weighted composite score with 5-tier classification. Top 14 secondary profiles scraped in medium and deep scans
- 🌍 Country detection — LLM identifies country of origin from profile signals with confidence % badge and on-demand per-commentor detection
- 🔬 Image intelligence — Vision LLM per photo for scene, crowd, symbols, weapons, OCR, and location estimation
- 🎬 Reel + text post intelligence — AI extracts entities, topics, narrative types, and threat indicators from captions and posts
- 👤 Face intelligence — HOG + CNN hybrid detection, 128D encoding, identity clustering, D3 face tree visualization
- 🕸️ Network graphs — Interactive D3 force graph, co-commentor heatmap matrix with shared posts on click, surname cluster edges
- 📄 PDF reports — Full intelligence report with image intelligence, reel intelligence, face gallery, post timeline, co-commentor analysis and multilingual script rendering
- 📬 Telegram delivery — PDF report sent automatically to configured chat on investigation completion
- 🤖 Local AI — Ollama powered, gemma4:e2b / e4b / 12b and other models, runs on GPU or CPU
- 🐳 Docker ready — One command deployment on Linux and Windows
⚠️ Disclaimer
BIRDY-EDWARDS is developed strictly for authorized intelligence, law enforcement, and academic research purposes only.
Scope of data access:
- This tool operates exclusively using a valid Facebook session authenticated by the operator
- It only accesses publicly visible profile data, posts, photos, reels, and comments
- It does not access private messages, locked profiles, restricted content, or any data not visible to a logged-in user
- It does not use bots, fake accounts, or automated account creation — the operator supplies their own authenticated session
Legal responsibility:
- This tool must only be used on profiles and content where you have explicit legal authorization to collect and analyze data
- Use without authorization may violate Facebook's Terms of Service, applicable privacy laws (GDPR, IT Act, DPDP Act), and local regulations
- The developer assumes no liability for misuse, unauthorized data collection, or any harm caused by improper use
- All investigations are the sole responsibility of the operator
By using BIRDY-EDWARDS, you confirm that your use is lawful, authorized, and compliant with all applicable laws in your jurisdiction.
System Requirements
| Component | Minimum | Recommended |
|---|
| OS | Ubuntu 24.04 LTS / Windows 10+ | Ubuntu 24.04 LTS |
| RAM | 8 GB | 16 GB |
| Storage | 20 GB free | 40 GB free |
| Docker | Docker Desktop / Engine | Latest stable |
| Ollama | Latest | Latest |
Installation
Prerequisites
Step 1 — Install Docker
Step 2 — Install Ollama
curl -fsSL https://ollama.com/install.sh | sh
Step 3 — Start Ollama bound to all interfaces
OLLAMA_HOST=0.0.0.0:11434 ollama serve
To make this permanent:
sudo systemctl edit ollama
Add:
[Service]
Environment="OLLAMA_HOST=0.0.0.0:11434"
sudo systemctl daemon-reload && sudo systemctl restart ollama
- Windows: Ollama listens on all interfaces by default — no extra configuration needed. If not then,
$env:OLLAMA_HOST="0.0.0.0"
ollama serve
Quick Start
Step 1 — Clone the repository
git clone https://github.com/jeet-ganguly/birdy-edwards.git
cd birdy-edwards
Step 2 — Create required files and directories
mkdir -p app/reports app/face_data app/post_screenshots app/status
touch app/fb_cookies.pkl app/socmint.db app/socmint_manual.db app/.ollama_model app/queue.db
jeet-ganguly/birdy-edwards
