apps-manage-payment-pages
Pass
Audited by Gen Agent Trust Hub on Sep 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests external data from the Apps platform via tools that list purchasers and subscribers. If these records contain attacker-controlled text, they could potentially influence the agent's behavior. 1. Ingestion points: Purchaser and subscriber list data retrieved via
apps_list_*tools (SKILL.md). 2. Boundary markers: Absent; no specific instructions are provided to treat record content as untrusted or to use delimiters. 3. Capability inventory: MCP tools for creating, updating, and deleting payment resources (apps_create_*,apps_delete_*). 4. Sanitization: Absent; the instructions direct the agent to summarize fields without performing validation or sanitization of the input strings.
Audit Metadata