inttegro-payment-refund
Pass
Audited by Gen Agent Trust Hub on Sep 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The instructions include specific security guardrails that prohibit the agent from requesting full account numbers, card details, or OTPs within the chat session, ensuring sensitive data remains out of the chat logs.
- [SAFE]: The skill utilizes an MCP tool provided by the vendor at
https://mcp.inttegro.com/for processing payments and refunds, which is an expected and authoritative source for this functionality. - [INDIRECT_PROMPT_INJECTION]: The skill processes order labels and refund descriptions retrieved from external tools. It implements mitigation by instructing the agent to provide a summary of all details and requiring final tool-level confirmation before committing any transaction.
Audit Metadata