sales-gainsight
Pass
Audited by Gen Agent Trust Hub on Apr 20, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill establishes a persistent knowledge-accumulation loop using references/learnings.md.\n
- Ingestion points: Untrusted user input regarding Gainsight tasks is processed for 'gotchas' and 'tips' which are then appended to references/learnings.md (Step 4).\n
- Boundary markers: The learnings file is read at the start of every session (Step 1) as 'accumulated platform knowledge' without any delimiters or instructions to ignore potentially malicious content within it.\n
- Capability inventory: The agent possesses file-read and file-append capabilities for the references/ directory. This persistent context is used to influence the agent's behavior and reasoning.\n
- Sanitization: There is no evidence of sanitization, filtering, or validation of the content being extracted from user interactions before it is committed to persistent storage.
Audit Metadata