analyzing-network-flow-data-with-netflow

Pass

Audited by Gen Agent Trust Hub on Jun 22, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the user to install the netflow package from a public registry to support NetFlow and IPFIX record parsing.
  • [COMMAND_EXECUTION]: The skill utilizes tcpdump for traffic capture and executes a local analysis script scripts/agent.py to process captured flows and generate reports.
  • [PROMPT_INJECTION]: The skill processes external network data, creating a surface for indirect prompt injection where malicious instructions could be embedded in flow metadata.
  • Ingestion points: Data enters the system context via captured_flows.json when parsed by the agent script.
  • Boundary markers: The instructions lack explicit delimiters or safety warnings to ignore instructions embedded in the analyzed data.
  • Capability inventory: The agent has the capability to execute shell commands and write files based on the analysis of these records.
  • Sanitization: No explicit sanitization or validation of the flow record content is described beyond standard library parsing.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 22, 2026, 10:00 AM
Security Audit — agent-trust-hub — analyzing-network-flow-data-with-netflow