handle-objection

Pass

Audited by Gen Agent Trust Hub on Nov 11, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external customer input, such as emails and transcripts, to analyze objections. This introduces a surface for indirect prompt injection. However, the skill employs structured templates and specific delimiters to manage this input, which is a standard requirement for its primary function.
  • [DATA_EXFILTRATION]: The skill generates links to Salesforce records using an instance URL template. These links are formatted to point to internal CRM records and do not represent an external exfiltration attempt.
  • [COMMAND_EXECUTION]: The skill uses authorized dispatch and dispatch_readonly tools to perform health checks and data queries against Salesforce APIs. These commands are scoped to standard CRM operations and are used to ensure tool connectivity before reporting errors to the user.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 11, 2026, 06:39 PM