AsadAliEngineer/RAG-Based-Threat-Intelligence-Assistant ? reverse-engineered prompt
Reverse engineered prompt
Build me a threat intelligence assistant that lets me ask natural language questions about CVEs, affected products, CWE categories, MITRE ATT&CK techniques, CAPEC patterns, and CISA KEV items, then gives me a clear answer with supporting context and prioritization help.
I want it to pull together vulnerability data from multiple sources, turn it into a searchable knowledge graph, and use semantic search plus graph relationships so the results feel smarter than a basic keyword lookup. The app should have a simple web interface for asking questions, a nice way to explore the relationships between threats and products, and a small analytics view that shows trends and coverage stats.
Also expose a FastAPI backend with endpoints for search, research answers, graph stats, and health checks, and make it work with local LLM inference if needed. If anything is unclear, look up current docs online and build the best practical version you can.
Are you gonna build this?
make sure you review the code using coderabbit