email-domain-security

Pass

Audited by Gen Agent Trust Hub on Aug 14, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides an embedded Python helper script (§10) that executes the system dig command via subprocess.run to perform recursive SPF record lookups.\n- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted external data in the form of DNS records, creating a vulnerability surface where malicious records can influence the agent or the diagnostic script.\n
  • Ingestion points: Raw DNS TXT records (SPF and DMARC) fetched from external domains.\n
  • Boundary markers: Not implemented; the skill instructions do not define delimiters to separate untrusted record content from the agent's instructional context.\n
  • Capability inventory: The skill enables shell command execution through the provided diagnostic script.\n
  • Sanitization: The helper script implements basic string normalization but lacks validation to ensure host strings derived from external records do not contain malicious command arguments (e.g., strings starting with a hyphen).
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 14, 2026, 11:32 AM
Security Audit — agent-trust-hub — email-domain-security