sap-unused-code

Pass

Audited by Gen Agent Trust Hub on May 12, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the SAPQuery tool with SAP_ALLOW_FREE_SQL=true to execute raw SQL queries constructed at runtime. While necessary for the audit functionality, this exposes a high-privilege execution path that could be misused if the platform does not enforce strict read-only constraints on the SQL runner.
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection via user-supplied filters (Package, Namespace, or Object list). These inputs are interpolated directly into SQL WHERE clauses (e.g., p~OBJ_NAME IN ( ... )) without instructions for sanitization or parameterized querying.
  • Ingestion points: User-provided filters described in the 'Required: scope and intent' section of SKILL.md.
  • Boundary markers: None present to distinguish user data from SQL command structure.
  • Capability inventory: SAPQuery executing arbitrary SQL based on user input.
  • Sanitization: No sanitization or validation logic is defined for the interpolated strings.
  • [EXTERNAL_DOWNLOADS]: The skill documentation references a public GitHub repository from Amazon Web Services (aws-solutions-library-samples) as a resource for parsing XML files. This is a reference to a well-known service and is documented as a fallback option for the user.
Audit Metadata
Risk Level
SAFE
Analyzed
May 12, 2026, 02:04 PM
Security Audit — agent-trust-hub — sap-unused-code