meta/agents/security-agent/skills/02-network-traffic-audit/SKILL.md
Monitor active outbound connections from all services and flag unexpected destinations
npx skillsauth add abzhaw/juliaz_agents network-traffic-auditInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Know exactly what this machine is talking to. Every outbound connection is either expected or suspicious. This skill makes that list visible every day.
api.openai.com — Julia's AI model
api.anthropic.com — Cowork MCP (Claude)
api.telegram.org — Telegram bot
smtp.gmail.com / imap.gmail.com — Email tools
medium.com — Julia medium agent
*.docker.com — Docker Hub pulls
localhost / 127.0.0.1 — Internal only
# Active outbound connections
lsof -i -n -P | grep ESTABLISHED
# DNS lookups (recent)
log show --predicate 'process == "mDNSResponder"' --last 1h 2>/dev/null | grep -i "query\|answer" | tail -50
| Finding | Severity | |---------|----------| | Connection to unknown IP | 🟠 High | | Process making unexpected external call | 🟠 High | | High volume of outbound requests | 🟡 Medium | | Known-good connection | ⚪ Info |
NETWORK AUDIT
✅ openai.com — orchestrator (expected)
✅ telegram.org — orchestrator (expected)
⚠️ Unknown: 185.220.101.x — process: node (PID 4521)
development
Fortschrittsverfolgung der Masterarbeit. Wortanzahl pro Kapitel, Fertigstellungsgrad, fehlende Elemente, Deadlines. Haelt den Ueberblick.
development
Kapitelarchitektur und Gliederung der Masterarbeit. Verwaltet die Struktur, schlaegt vor wo Inhalte hingehoeren, validiert den logischen Fluss zwischen Kapiteln.
tools
Konvertiert Protokolleinträge und Session-Logs in thesis-fähiges deutsches Narrativ. Transformiert Entwicklungsdokumentation in akademische Prosa.
research
Sucht und analysiert akademische Literatur. Findet relevante Papers, erstellt strukturierte Zusammenfassungen. Zitiert NIEMALS — schlaegt nur vor.