sap-btp-job-scheduling

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides numerous examples of shell commands using curl and the Cloud Foundry (cf) CLI for service management, authentication, and job scheduling operations (documented in references/rest-api.md, references/setup-guide.md, and references/security.md). These are standard administrative actions for the target platform.\n- [EXTERNAL_DOWNLOADS]: The skill references and provides instructions for installing official SAP Node.js packages including @sap/jobs-client, @sap/xsenv, and @sap/xssec from the public npm registry, as seen in SKILL.md and references/rest-api.md. These are verified libraries for SAP BTP development.\n- [INDIRECT_PROMPT_INJECTION]: The skill guides the agent in generating and potentially processing job configurations and action endpoint responses. This presents a surface for indirect prompt injection if the agent interacts with untrusted external data during job execution or log analysis.\n
  • Ingestion points: The skill uses templates (templates/job-creation.json) and service responses (run logs) described in references/rest-api.md.\n
  • Boundary markers: No explicit delimiter or security warnings for the agent regarding the separation of instructions and external data are present in the core instructions in SKILL.md.\n
  • Capability inventory: The skill utilizes network operations via curl, management via cf CLI, and file-based configurations across references/rest-api.md and references/setup-guide.md.\n
  • Sanitization: Input validation instructions are limited to basic character restrictions for job names in references/concepts.md and references/rest-api.md.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 04:22 AM
Security Audit — agent-trust-hub — sap-btp-job-scheduling