┌─────────────────────────────────────────────────────────────────────────┐
│ I build autonomous systems. │
│ Security agents, OSINT pipelines, and local AI that investigate, │
│ decide, and act — without supervision. │
└─────────────────────────────────────────────────────────────────────────┘
class AnubhavMohandas:
location = "Kerala, India"
active_since = 2019
current_focus = ["Autonomous Security Agents", "OSINT Pipelines", "Local AI Systems"]
expertise = {
"autonomous_ai": ["Security Agents", "OSINT Automation", "Local LLMs"],
"forensics": ["Memory Analysis", "Disk & Network Forensics", "Cyber Crime Investigation"],
"offensive": ["Penetration Testing", "Payload Generation", "Exploit Research"],
"defensive": ["SIEM Engineering", "Kernel-level Threat Detection", "Honeypots"],
"intelligence": ["OSINT", "Threat Intelligence", "Reconnaissance", "Attribution"],
}
currently_building = "Systems that investigate, decide, and act — without supervision."Click any system to expand ↓
◈ SAGE — Security Analysis & Graph Engine 
CVE drops. SAGE wakes up.
Maps your entire transitive dependency graph using tree-sitter + NetworkX, finds every function in the blast radius, calls Claude for a targeted patch, runs CVE-specific exploit tests to verify it, then opens a GitHub PR automatically.
You review and merge. That's it.
tree-sitter · NetworkX · Claude API · GitHub API
◈ NYX — Personal AI Assistant
Fully local. No API keys phoning home. Voice-activated, multi-model fallback.
Speaks Hindi / English / Hinglish / Malayalam — because your AI should understand how you actually think.
The anti-cloud AI assistant.
Local LLMs · Whisper · Python · Multi-model fallback
◈ MIDAS — Market Intelligence, Discovery & Autonomous SaaS-builder
Goes to sleep. Wakes you up with a live product.
Mines Reddit / Twitter / ProductHunt / App Store for pain signals → validates willingness-to-pay → scaffolds FastAPI + Supabase + Stripe → ships to Vercel → posts to ProductHunt.
Overnight. Autonomous. No human in the loop.
FastAPI · Supabase · Stripe · Vercel · Python
◈ CyberTrace — OSINT Intelligence Platform
One identifier in. Everything out.
| Input | Output |
|---|---|
Email |
Breach history, associated accounts |
BTC Address |
Wallet graph, transaction trail |
Phone |
Carrier, owner, linked profiles |
IP |
Geolocation, Shodan exposure |
GSTIN |
Full business intelligence |
30+ sources across surface, deep, and dark web — queried in parallel, results merged and ranked.
Python · Shodan API · 30+ OSINT sources · Parallel querying
◈ HERMES — Hybrid Engine for Research, Memory, Execution & Synthesis
54 open-source patterns. 20 modules. 5 layers. One install.
Built so Claude can do everything without switching contexts — research, write, remember, execute, synthesize.
Python · Claude API · MCP · Custom memory layer
| Tool | What it actually does |
|---|---|
| 🔐 GHOST | AES-256-GCM autofill extension. Reverse-engineered IRCTC's Angular monolith for Tatkal booking. No public API used. |
| 📄 Nyxine | Resume builder that never touches a server. JD keyword scoring, ATS gap analysis, AI coaching — all in-browser. |
| 🔍 WhoisUser | Username recon across 100+ platforms. One query. |
| 💣 Secure Gen | Payload framework — 15+ types, WAF bypass, DB-specific vectors. Ethical hacking only. |
| 🛡️ SIEM Exploit Detector | Real-time APT and kernel-level exploit detection. |
| 🕵️ Digital Forensics Toolkit | All-in-one investigation suite for cyber crime cases. |
| 📊 Enhanced CVSS Calculator | Because the standard one isn't enough for real triage. |
| 🕸️ Web Honeypot | Attacker behavior capture and analysis in real-time. |
| 📋 Log Analyzer | Advanced threat detection engine. |
Languages & Frameworks
AI & Autonomous Systems
Security Arsenal
Forensics & OS
$ curl -X GET https://anubhavmohandas.netlify.app
→ Security-first. Autonomous. Zero toolchain overhead.
┌──────────────────────────────────────────────────────────────────────┐
│ "Offense informs defense. Know your adversary." │
│ │
│ Open for collaboration on security research & autonomous AI │
│ With greater power comes greater responsibility │
└──────────────────────────────────────────────────────────────────────┘
security · forensics · autonomous systems · 2019 → now