inbound-lead-triage

Pass

Audited by Gen Agent Trust Hub on Jun 23, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection by processing untrusted lead data from CSV files that is intended for later consumption by an AI agent.
  • Ingestion points: scripts/triage.py ingests data from CSV files specified via the --inputs command-line argument.
  • Boundary markers: The skill does not implement boundary markers or instructions to alert the downstream agent to ignore potential instructions embedded within lead data fields.
  • Capability inventory: The skill includes file reading (scripts/triage.py) and file writing (JSON queue output), but does not possess network access or remote code execution capabilities.
  • Sanitization: No sanitization or validation is applied to the content of CSV fields (e.g., company, name) before they are included in the triage queue.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 23, 2026, 04:36 PM
Security Audit — agent-trust-hub — inbound-lead-triage