configure
Pass
Audited by Gen Agent Trust Hub on Nov 11, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses a Python script to perform text substitution in an HTML template. This is a local file operation used to inject user-specific Salesforce identifiers (User ID, Tool Name) into a dashboard artifact. The script specifically resolves local plugin directories and does not accept or execute untrusted external input.
- [DATA_EXPOSURE_EXFILTRATION]: The skill queries standard Salesforce user identity and pipeline data (Opportunities, Leads, Tasks). This data is used solely to populate a local UI artifact (
Your Sales Intelligence) within the agent platform. No network exfiltration to third-party domains is detected; all API calls are directed to the authenticated Salesforce instance via authorized tools. - [REMOTE_CODE_EXECUTION]: All dashboard logic is contained within a local HTML template. The dashboard interacts with Salesforce using the
window.cowork.callMcpToolbridge, which is restricted to a single verified tool name defined at build time. It does not load external scripts or styles from remote servers. - [PROMPT_INJECTION]: The skill contains instructions for the agent to maintain a specific persona and silent execution style, which are standard for user-experience consistency and do not attempt to bypass safety filters or ignore system instructions.
Audit Metadata