account-brief

Fail

Audited by Gen Agent Trust Hub on Aug 4, 2026

Risk Level: CRITICALPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill's author metadata identifies 'Builder.org' as the creator. This domain has been explicitly flagged as malicious and blacklisted by automated security scanners, indicating that the skill originates from a potentially harmful source.
  • [PROMPT_INJECTION]: The skill has an indirect prompt injection surface through its ingestion of untrusted external data during the briefing workflow.
  • Ingestion points: Processes external web research (news, funding, leadership changes) and HubSpot CRM records (SKILL.md).
  • Boundary markers: The workflow lacks explicit delimiters or instructions to the agent to ignore embedded commands or adversarial instructions within the retrieved web data.
  • Capability inventory: Utilizes web search tools and HubSpot read capabilities.
  • Sanitization: The instructions provide no evidence of filtering, escaping, or validation of external content before it is processed by the model.
Recommendations
  • Contains 1 malicious URL(s) - DO NOT USE
Audit Metadata
Risk Level
CRITICAL
Analyzed
Aug 4, 2026, 04:37 PM
Security Audit — agent-trust-hub — account-brief