vip-alert
Pass
Audited by Gen Agent Trust Hub on Jul 1, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements a standard business workflow using the author's own 'composio' CLI and toolkits. All data access (Gorgias tickets, HubSpot profiles) and network operations (Slack alerts) are consistent with the described purpose of VIP monitoring.
- [PROMPT_INJECTION]: Indirect Prompt Injection Surface: The skill processes external content from support tickets and CRM records which could contain adversarial instructions designed to influence the agent's output.
- Ingestion points: Ticket data retrieved via GORGIAS_LIST_TICKETS and contact data from HUBSPOT_SEARCH_CONTACTS_BY_CRITERIA in Step 2 and 3 of SKILL.md.
- Boundary markers: Absent. The skill does not explicitly use delimiters to isolate untrusted data from instructions.
- Capability inventory: The agent has the capability to post messages to Slack via SLACK_SEND_MESSAGE and execute further tool commands via the 'composio' CLI.
- Sanitization: Absent. No explicit filtering or validation of ticket/CRM content is mentioned.
- Note: This is documented as an inherent architectural surface for skills processing external data; the risk is considered low given the professional contexts of the tools used.
Audit Metadata