skills/serac-labs/serac/inbound-email/Gen Agent Trust Hub

inbound-email

Warn

Audited by Gen Agent Trust Hub on Sep 6, 2026

Risk Level: MEDIUMDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [DYNAMIC_EXECUTION]: The skill instructs the agent to use the snow_record_manage tool to write raw JavaScript code into the script column of the sysevent_in_email_action table. This methodology results in the runtime generation and execution of scripts within the ServiceNow platform.
  • [INDIRECT_PROMPT_INJECTION]: The skill documents how to ingest untrusted data from external email subjects and bodies into ServiceNow records, creating a vulnerability surface. 1. Ingestion points: Data entering the system via the sys_email table (specifically email.subject and email.body_text). 2. Boundary markers: Absent; there are no instructions to use delimiters or ignore instructions within the ingested content. 3. Capability inventory: The skill utilizes snow_record_manage for database writes and snow_execute_script for testing script logic. 4. Sanitization: Absent; the example code directly maps external email strings to record fields without validation or escaping.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Sep 6, 2026, 07:15 PM
Security Audit — agent-trust-hub — inbound-email