sf-datacloud-retrieve

Pass

Audited by Gen Agent Trust Hub on Mar 22, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes local scripts and system commands to perform diagnostics and data operations within Salesforce orgs.
  • Evidence: The skill invokes a local script using node ~/.claude/skills/sf-datacloud/scripts/diagnose-org.mjs (SKILL.md).
  • Evidence: Multiple calls to the sf data360 CLI are used for SQL queries, describes, and search-index management (README.md, SKILL.md).
  • [EXTERNAL_DOWNLOADS]: The skill requires an external community-developed plugin for the Salesforce CLI to function.
  • Evidence: The compatibility section in SKILL.md specifies: "Requires an external community sf data360 CLI plugin".
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it ingests and processes data from external Salesforce Data Cloud environments.
  • Ingestion points: Data retrieved through sf data360 query sql, sf data360 query describe, and sf data360 search-index list (SKILL.md).
  • Boundary markers: No delimiters or explicit instructions to ignore embedded commands are present in the provided operational rules.
  • Capability inventory: The skill has the ability to execute system commands via the sf CLI and run Javascript files via node (SKILL.md).
  • Sanitization: There is no evidence of output validation or sanitization before the agent interprets retrieved data (SKILL.md).
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 22, 2026, 09:20 PM