win-loss-review
Pass
Audited by Gen Agent Trust Hub on Nov 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted data from external sources, specifically call transcripts and email threads, to identify qualitative deal patterns.
- Ingestion points: External data enters the agent context through the use of doc and email tools in the evidence gathering phase (Step 2b).
- Boundary markers: The instructions do not define explicit delimiters or use 'ignore-internal-instructions' markers when handling the content of transcripts or emails.
- Capability inventory: The skill possesses the capability to query Salesforce data via
dispatch_readonlyand generate user-facing output viadisplay_widget. - Sanitization: There are no specified sanitization or validation steps for the external text data before it is interpolated into the agent's reasoning process.
Audit Metadata