handle-complaint
Pass
Audited by Gen Agent Trust Hub on Sep 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted customer messages loaded from external sources, which is a potential surface for indirect prompt injection attacks.
- Ingestion points: Content is loaded from email (
~~почта) and CRM tickets (~~crm) in SKILL.md (Step 1). - Boundary markers: There are no explicit delimiters or instructions to ignore embedded commands within the ingested text.
- Capability inventory: The skill is configured with
allowed-tools: Read, WebFetch, Bash. - Sanitization: No sanitization, escaping, or validation of the external customer input is specified.
- [SAFE]: The skill does not contain any detected malicious patterns, obfuscation, or unauthorized access to sensitive file paths. The risk of processing external data is mitigated by strict "Approval gates" that require a human to review and approve all responses and actions before they are executed.
Audit Metadata