attribution
Pass
Audited by Gen Agent Trust Hub on Sep 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from various marketing and CRM platforms (HubSpot, Salesforce, PostHog, etc.) to reconcile conversion numbers. Ingesting this untrusted external data creates a potential surface for indirect prompt injection, where malicious instructions could be embedded in tool outputs.
- Ingestion points: Data from connected CRM and analytics tools (e.g., pipeline reports, attribution data, user profiles) described in SKILL.md and integration guides.
- Boundary markers: The skill lacks explicit instructions for the agent to treat tool-provided text as untrusted or to use specific delimiters when processing these inputs.
- Capability inventory: The skill is designed to work with tools for reading from and writing to sensitive data systems (CRM/Analytics).
- Sanitization: No explicit sanitization or validation of external tool content is specified for the agent.
Audit Metadata