netflows

Warn

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [Unverifiable Dependencies] (MEDIUM): The skill relies on a binary named 'netflows' located in the 'iothackbot bin directory'. This source is not from a trusted provider, and the integrity or safety of the binary cannot be verified from the provided metadata.
  • [Indirect Prompt Injection] (LOW): The skill is designed to analyze external pcap/pcapng files, which are untrusted data sources. Malicious packet data (e.g., DNS response strings) could be used to influence the agent's interpretation of the network flows.
  • Ingestion points: Files processed via the 'pcap_files' parameter in SKILL.md.
  • Boundary markers: Absent. The agent is instructed to 'Interpret results' and 'Explain significance' without clear delimiters or instructions to ignore embedded commands.
  • Capability inventory: Execution of the 'netflows' command. No network-write or file-write capabilities are explicitly requested by the skill, limiting the risk to the agent's internal reasoning and summary output.
  • Sanitization: Absent. The output from the tool is parsed and presented to the agent without escaping or filtering.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 16, 2026, 04:27 AM