connect-hubspot-mcp
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill enables an attack surface for indirect prompt injection by instructing the agent to process untrusted data from an external CRM.
- Ingestion points: The agent reads CRM records including contacts, companies, deals, tickets, campaigns, landing pages, and blog posts from HubSpot via
mcp.hubspot.com(as described in SKILL.md). - Boundary markers: No explicit delimiters or instructions are provided to help the agent distinguish between legitimate data and potential embedded instructions within the CRM fields.
- Capability inventory: The agent is granted capabilities to read, write, and create CRM objects and website content.
- Sanitization: The skill does not provide methods for sanitizing or filtering the external content before it is processed by the agent.
Audit Metadata