zendesk-triage
Pass
Audited by Gen Agent Trust Hub on Jun 16, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
composioCLI to discover tools, fetch schemas, and execute Zendesk API calls during the ticket triage process.- [PROMPT_INJECTION]: The skill ingests untrusted data from Zendesk tickets, creating a surface for indirect prompt injection. - Ingestion points: Ticket details and comments fetched in Step 4 of
SKILL.md. - Boundary markers: Absent; ticket content is interpolated into the triage context without delimiters.
- Capability inventory: Subprocess execution via
composiofor searching and fetching data. - Sanitization: Absent; ticket content is processed directly for priority and category classification.
Audit Metadata