experience-ui-bundle-salesforce-data-access

Pass

Audited by Gen Agent Trust Hub on Aug 14, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes several CLI tools for developer workflows, including node, npm, npx, and the Salesforce CLI (sf). It specifically employs graphiti (a Salesforce query compiler) to generate guardrail-applied GraphQL queries and a local bash script graphql-search.sh to search the project's schema.graphql file for entity metadata.
  • [EXTERNAL_DOWNLOADS]: The skill requires the installation of @salesforce/platform-sdk and @salesforce/graphiti. These are official, vendor-owned packages from the skill author (Salesforce) used for authenticating and interacting with the Salesforce GraphQL and REST APIs.
  • [DATA_EXPOSURE]: The skill processes Salesforce record data and metadata (schema.graphql). These operations are performed through official SDKs that handle authentication and CSRF protection, targeting authorized Salesforce environments.
  • [INDIRECT_PROMPT_INJECTION]: The skill defines an attack surface by ingesting data from Salesforce APIs (records and metadata) and possessing command-line capabilities via the agent. However, as the skill is authored by the platform vendor for developer use cases and provides specific templates for safe data handling, the risk is negligible.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 14, 2026, 08:39 PM
Security Audit — agent-trust-hub — experience-ui-bundle-salesforce-data-access