hubspot-agency-multi-portal

Pass

Audited by Gen Agent Trust Hub on Sep 9, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The implementation guide and onboarding scripts use subprocess.run to interact with system utilities including secret managers (pass, aws), version control tools (gh), and system services (systemctl, kill).
  • [INDIRECT_PROMPT_INJECTION]: The skill provides scripts that process external data files, creating a surface for indirect injection attacks.
  • Ingestion points: The bulk-onboard.py script reads user-provided CSV files, and compliance-report.py processes JSONL audit logs.
  • Boundary markers: The skill does not implement explicit boundary markers or delimiters to isolate processed data from agent instructions.
  • Capability inventory: Scripts within the skill possess capabilities for network communication (HubSpot API), file system modifications, and execution of shell commands via subprocesses.
  • Sanitization: The scripts rely on standard parsing (CSV/JSON) without advanced sanitization of the values retrieved from external files.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 9, 2026, 03:47 AM
Security Audit — agent-trust-hub — hubspot-agency-multi-portal