customer-success-manager
Pass
Audited by Gen Agent Trust Hub on Feb 26, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [NO_CODE]: The skill is entirely composed of a markdown file (SKILL.md) that outlines the agent's persona and communication protocols. No executable scripts, binary files, or configuration for remote code execution are present.
- [SAFE]: No malicious patterns such as hardcoded credentials, data exfiltration, or persistence mechanisms were found. The skill does not attempt to access sensitive file paths or initiate network connections.
- [PROMPT_INJECTION]: The skill processes external customer data retrieved from a context manager, which is a potential surface for indirect prompt injection. However, this poses no threat as the skill has no dangerous capabilities (such as shell access, file writing, or network tools) that an attacker could exploit.
- Ingestion points: Customer data accessed via the 'get_customer_context' request.
- Boundary markers: None present in the prompt instructions.
- Capability inventory: None. No system-level tools or subprocess calls are defined.
- Sanitization: No sanitization or validation of the input data is described.
Audit Metadata