acceptance-criteria

Pass

Audited by Gen Agent Trust Hub on Aug 3, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes the Bash native tool to perform REST API operations (GET and PATCH) against ServiceNow table endpoints for reading and updating story records.
  • [PROMPT_INJECTION]: The skill processes external data from ServiceNow records to generate acceptance criteria, which introduces an indirect prompt injection surface.
  • Ingestion points: Data is retrieved from the description, short_description, and acceptance_criteria fields of the rm_story and rm_epic tables (Step 1, Step 2, and Step 7).
  • Boundary markers: Absent; the instructions do not implement delimiters or warnings to ignore instructions that might be embedded in the retrieved ServiceNow data.
  • Capability inventory: The skill has write access via the SN-Update-Record tool and Bash-based REST PATCH requests, allowing it to modify records based on processed data.
  • Sanitization: No explicit sanitization or validation of the ingested record content is performed before interpolation into the generation prompt.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 3, 2026, 02:24 PM
Security Audit — agent-trust-hub — acceptance-criteria