plaid-error-handling
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data returned from an external source (Plaid API), which introduces an indirect prompt injection surface.
- Ingestion points: Error objects and messages are ingested from API responses via
error.response.dataas shown in the code snippets inSKILL.md. - Boundary markers: The instructions do not specify the use of delimiters or warnings to ignore potential instructions embedded in technical error fields.
- Capability inventory: The skill facilitates database updates (
db.items.update), user notifications (notifyUser), and additional API calls (plaidClient.linkTokenCreate,plaidClient.transactionsSync). - Sanitization: No validation or sanitization logic is provided for the technical error messages (
error_message) before they are logged or processed.
Audit Metadata