sf-industry-commoncore-integration-procedure
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The analyzed skill implements a highly structured process for generating Salesforce Integration Procedures. It prioritizes security by including a 110-point scoring system and mandatory guardrails that explicitly block dangerous patterns like hardcoded credentials or unvalidated DML operations.\n- [COMMAND_EXECUTION]: The skill uses official Salesforce CLI tool commands (e.g.,
sf data query,sf project deploy) to manage metadata and project resources. These operations are transparent, well-documented, and aligned with standard Salesforce development workflows.\n- [INDIRECT_PROMPT_INJECTION]: While the skill processes user-supplied requirements to generate code equivalents, it incorporates defensive documentation and prompting techniques. It defines a multi-phase workflow with strict validation rules that check for circular dependencies, unbounded queries, and missing sanitization in the resulting output. (Ingestion: Phase 1 Requirements; Boundary: Phase 3 Validation Guardrails; Capability: sf CLI deployment; Sanitization: Security scoring criteria).
Audit Metadata