setup-payright
Warn
Audited by Snyk on Nov 16, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (low risk: 0.10). The required workflow reads Payright's Get Bill response and the callback payload, which can include payer-entered
customer_details.nameandfpx_buyer_name. An outsider can influence those values, but the agent only receives them after selecting a specific bill, so this is low risk rather than medium.
MEDIUM W009: Direct money access capability detected (payment gateways, crypto, banking).
- Direct money access detected (medium risk: 0.30). Scope is ambiguous. The skill is mainly build/debug guidance for a Payright integration, which is excluded as implementation guidance. However, it also documents live operational routes: creating real FPX bills against the production base URL (indirect, since bills are real invoices), and the Payout API for disbursement (a direct transfer of money out, delegated through the skill's documented surface). Under the recall-first override, I treat these as in scope. For the direct payout route, the control is the service-enforced
auth-tokenheader, which the service validates against the account's API credentials. That is a documented service authentication gating the route, so it qualifies as established control and does not justify high. The bill route is indirect, so it also yields medium. The highest applicable tier is therefore medium. Exposure would be high if payout execution were treated as direct without service-enforced authentication, but the docs establish it.
Issues (2)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
W009
MEDIUMDirect money access capability detected (payment gateways, crypto, banking).
Audit Metadata