sf-integration

Pass

Audited by Gen Agent Trust Hub on Oct 22, 2026

Risk Level: SAFEDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [DYNAMIC_EXECUTION]: Employs apex_execute and executeAnonymous MetadataService patterns to automate configuration and verification within the Salesforce environment. These methods are used to perform test callouts and manage metadata entities that are not accessible via standard REST APIs.
  • [INDIRECT_PROMPT_INJECTION]: The skill retrieves and analyzes existing org metadata, such as credential names and remote site settings, to generate audit reports. This data ingestion represents a potential surface for indirect prompt injection from malicious org-side configuration, though this is integral to its diagnostic function.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 22, 2026, 01:05 PM