draft-reply

Pass

Audited by Gen Agent Trust Hub on Jul 1, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill has a surface for indirect prompt injection because it ingests untrusted data from an external support system.
  • Ingestion points: Untrusted conversation threads and message details are fetched via the GORGIAS_GET_TICKET tool in SKILL.md.
  • Boundary markers: No explicit delimiters or instructions are used to separate the ticket data from the system instructions.
  • Capability inventory: The skill has the capability to write data by creating drafts in Gmail via GMAIL_CREATE_EMAIL_DRAFT in SKILL.md.
  • Sanitization: There is no evidence of sanitization or validation of the fetched ticket content before it is analyzed by the agent.
  • [COMMAND_EXECUTION]: The skill uses the composio CLI to search for tools, fetch schemas, and execute API calls, which is standard behavior for the vendor.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 1, 2026, 08:13 PM
Security Audit — agent-trust-hub — draft-reply