hubspot-core-workflow-a

Pass

Audited by Gen Agent Trust Hub on Mar 25, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill uses the official and well-known @hubspot/api-client package to interact with HubSpot services.
  • [SAFE]: Access tokens are managed via environment variables (process.env.HUBSPOT_ACCESS_TOKEN), which is a standard security practice for credential management.
  • [PROMPT_INJECTION]: The skill contains an indirect prompt injection surface as it processes external lead data (emails, names, company names) into HubSpot records.
  • Ingestion points: Ingests untrusted data through the LeadInput interface properties (e.g., firstName, lastName, email).
  • Boundary markers: Absent; input data is passed directly into the HubSpot SDK client methods.
  • Capability inventory: The skill performs API operations to search, create, and update contacts, companies, deals, and notes in the HubSpot CRM.
  • Sanitization: No string sanitization or validation is applied to the lead input data before it is sent to the HubSpot API.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 25, 2026, 04:41 PM