migrate-segw-to-rap

Pass

Audited by Gen Agent Trust Hub on Jun 22, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests untrusted data in the form of existing ABAP source code and uses it to drive code generation.
  • Ingestion points: The skill reads ABAP class source code (MPC and DPC_EXT) in Phases 2 and 3 using the SAPRead tool.
  • Boundary markers: The instructions do not specify the use of boundary markers or instructions for the agent to ignore embedded commands within the source code being analyzed.
  • Capability inventory: The skill possesses extensive capabilities across all scripts, including file writing (SAPWrite), object activation (SAPActivate), database queries (SAPQuery), and network operations (curl).
  • Sanitization: No sanitization, escaping, or validation of the ingested source code is described before the content is interpolated into prompts for generating new RAP services.
  • [COMMAND_EXECUTION]: The skill utilizes curl commands in Phase 7 to perform smoke tests and verify data parity between legacy and new services.
  • Evidence: The skill executes network operations such as curl -u "$USER:$PASS" "<base>/sap/opu/odata/sap/<legacy_service>/ProjectSet/\$count" and handles session-sensitive data like CSRF tokens and cookies.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 22, 2026, 11:09 AM
Security Audit — agent-trust-hub — migrate-segw-to-rap