Network tested inside and out, then watched every week
I tested the office network from inside and outside, checked the findings and ranked the fixes by risk. A weekly scan reports new devices and changed ports.
Results
4phishing campaigns traced and blocked
24 of 24checked, with no matching worm indicators found
2large customers' replies out of quarantine
The problem
I needed to identify exposed devices and check what an outsider could reach. Phishing emails targeted the finance inbox, while some large customers' filters quarantined sales replies without sending a bounce.
What I did
- Ran vulnerability assessments from a test machine inside the network and a throwaway cloud server outside it, then ranked the findings by risk.
- A weekly scan that compares devices and open ports against a baseline and posts what changed, matching hardware addresses to cut false alarms.
- Traced each phishing campaign through its redirects and look-alike domains, blocked what it used, and added DNS filtering for the office.
- Swept the fleet for a self-spreading npm worm the week it surfaced, and added a 30-day wait before adopting new package versions.
- Found outgoing mail was signed only with the provider's shared domain, then published the company's own signing key and confirmed it in message headers.
What I took from it
Some serious-looking findings were false alarms when I checked them again. I verified them before putting them in the remediation plan, so the fixes were based on confirmed problems.