Organizing recon output with AI
Organizing recon output with AI means handing a model the hostnames and notes you already saved, and getting them back grouped by asset and scope. The model sorts text; it sends no requests.
Split the work
| Give the model | Keep for yourself |
|---|---|
| Group hostnames under each scope rule | Deciding an edge case; ask the program |
| Merge notes from several tools and drop duplicates | Choosing what to test next |
| List the technologies your notes mention | Sending any request to a target |
| Flag hosts that match no scope rule | Confirming that anything is a finding |
| Turn notes into a to-do list per asset | Respecting rate and automation limits |
From the shell
Pipe a saved file into a model with LLM CLI.
# scope.md holds your checked scope list, hosts.txt your saved hostnamescat scope.md hosts.txt | llm -m flash 'Group these hostnames under the scope rules at the top. Mark any host that matches no rule as UNKNOWN. Do not add hosts.'Automation limits still apply
"Be gentle when conducting automated tests or scanners. Some programs may disallow automated testing of any kind, or impose rate limits."
Intigriti Community Code of Conduct, read November 18, 2026
Redact session tokens, cookies and other users' data before you paste notes anywhere. GLM-5.3 Flash and DeepSeek V4.1 Flash 1017 both cost $0.24 per 1M input tokens, which suits long lists.