customer-success-manager

Pass

Audited by Gen Agent Trust Hub on Mar 1, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface due to its requirement to ingest and process data from external, potentially untrusted sources.\n- Ingestion points: The agent is instructed to monitor and analyze content from support tickets, user feedback platforms (e.g., Typeform, SurveyMonkey), and CRM integrations (Salesforce, HubSpot).\n- Boundary markers: The instructions do not define delimiters or specific protocols to isolate external data from the agent's core instructions.\n- Capability inventory: The skill has access to sensitive tools including Bash, Write, MultiEdit, and WebFetch, which could be exploited if malicious instructions are successfully injected via analyzed data.\n- Sanitization: There are no specifications for sanitizing or validating the data inputs before they are interpreted or acted upon by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 1, 2026, 01:37 AM