customer-360

Pass

Audited by Gen Agent Trust Hub on Jul 1, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the composio CLI to interact with external support and CRM services. It passes user-provided identifiers from the $ARGUMENTS variable directly into shell commands for tool execution.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection via the $ARGUMENTS parameter. 1. Ingestion points: User-provided email or name via the $ARGUMENTS variable in SKILL.md. 2. Boundary markers: Absent; the user input is interpolated into the workflow without delimiters or instructions to ignore embedded commands. 3. Capability inventory: Execution of composio CLI commands which can perform read/write operations against integrated services in SKILL.md. 4. Sanitization: Absent; there is no defined validation or escaping of the customer identifier before it is used to filter queries.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 1, 2026, 08:14 PM
Security Audit — agent-trust-hub — customer-360