close-plan

Pass

Audited by Gen Agent Trust Hub on Nov 11, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill uses established Salesforce GraphQL endpoints (/services/data/v65.0/graphql) and follows best practices for data grounding by querying object metadata before relying on custom fields.
  • [SAFE]: No remote code execution or unauthorized external network calls were detected. All network operations are directed through the authorized Salesforce and Headless 360 connector tools (dispatch_readonly, dispatch).
  • [SAFE]: The skill implements strict data isolation rules, such as failing fast on empty MINE scopes to prevent accidental data leakage from other users' records.
  • [SAFE]: Output formatting is handled via a structured widget template or markdown fallback, using safe interpolation of variables without executable script generation.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from transcripts, emails, and Slack. It mitigates risk by requiring external knowledge (value props, approval chains) to be grounded in the org's own data and specifically instructions the agent not to invent data, though standard precautions for processing third-party text should always be observed.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 11, 2026, 06:39 PM