platform-data-manage

Pass

Audited by Gen Agent Trust Hub on Aug 27, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided data to generate and execute Salesforce data operations, which is its primary purpose.
  • Ingestion points: User requests for record creation, updates, or queries are processed via sf CLI and anonymous Apex scripts, as defined in SKILL.md.
  • Boundary markers: While explicit instruction boundary markers are not used in the interpolated prompts, the skill mandates a 'describe-first' validation workflow to ensure data matches the target org's schema.
  • Capability inventory: The skill utilizes sf data and sf apex run tools for record management and server-side logic execution in Salesforce environments.
  • Sanitization: The skill includes local validation scripts (scripts/soql_validator.py and scripts/validate_data_operation.py) to analyze and score the safety and efficiency of data operations before they are performed.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 27, 2026, 04:35 PM
Security Audit — agent-trust-hub — platform-data-manage