I Built a Fully Local OSINT Agent with Ollama, LangChain, Telegram and Qwen3.5 14B — Running 24/7 on My Homelab, Zero Cloud, Zero Compromises

📰 Dev.to · miwaty

Learn to build a fully local OSINT agent using Ollama, LangChain, Telegram, and Qwen3.5 14B, running 24/7 on your homelab without relying on the cloud

advanced Published 12 Apr 2026
Action Steps
  1. Install Ollama and LangChain on your local machine to create the foundation for your OSINT agent
  2. Configure Telegram to act as the communication interface for your agent
  3. Integrate Qwen3.5 14B to provide the necessary language model capabilities
  4. Set up your homelab to run the agent 24/7, ensuring continuous operation and monitoring
  5. Test and refine your agent to ensure it meets your specific OSINT needs and requirements
Who Needs to Know This

This project is ideal for a team of developers and security researchers who want to create a self-contained OSINT agent for their organization, allowing them to maintain full control over their data and operations

Key Insight

💡 By building a local OSINT agent, you can maintain full control over your data and operations, reducing reliance on cloud services and potential security risks

Share This
🚀 Build a fully local OSINT agent with Ollama, LangChain, Telegram, and Qwen3.5 14B, no cloud required! 🤖
Read full article → ← Back to Reads