omnistudio-callable-apex-generate
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is designed for static analysis and generation of Salesforce Apex code. It includes a comprehensive 120-point scoring rubric that prioritizes security, specifically targeting CRUD (Create, Read, Update, Delete) and FLS (Field-Level Security) checks. The provided code patterns (e.g., in
assets/pattern_callable_vanilla.cls) explicitly demonstrate the use ofWITH USER_MODEandwith sharingto ensure secure data access within the Salesforce platform. - [INDIRECT_PROMPT_INJECTION]: The skill has a defined attack surface as it is instructed to read and review existing local Apex files (e.g.,
Glob: **/*Callable*.cls). - Ingestion points: Reads existing callable class files during Phase 1 of its workflow as specified in
SKILL.md. - Boundary markers: None explicitly defined in the instructions for separating user code from agent instructions during the review process.
- Capability inventory: The skill provides review commentary and generates new Apex source code. It does not possess capabilities for direct network transmission or shell command execution; deployment and testing are delegated to the
platform-metadata-deployandplatform-apex-test-runskills. - Sanitization: Not present. However, because this code review functionality is the primary intended purpose of the skill, and the skill does not execute the code it reviews, this is considered a standard operational risk handled by the platform's underlying model guardrails.
Audit Metadata