weekly-wrap
Pass
Audited by Gen Agent Trust Hub on Nov 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from Salesforce opportunity fields, email threads, and Slack messages to generate its weekly recap.
- Ingestion points: Salesforce GraphQL/SOQL queries (Step 2), email tools, and Slack/document connector tools.
- Boundary markers: Absent. There are no instructions to delimit or treat external data as untrusted text within the agent's context.
- Capability inventory: Access to
dispatch,dispatch_readonly, anddisplay_widgettools. - Sanitization: Absent. The instructions focus on data accuracy but do not include security sanitization or escaping of the ingested content before it is used in the UI or summary.
- [SAFE]: The skill interacts exclusively with official Salesforce API endpoints and internal messaging tools. No evidence of data exfiltration to unauthorized third-party domains or hardcoded credentials was found.
Audit Metadata