setup-setel

Fail

Audited by Snyk on Nov 16, 2026

Risk Level: CRITICAL
Full Analysis

CRITICAL E004: Prompt injection detected in skill instructions.

  • Potential prompt injection detected (medium risk: 0.30). The skill is a documented Setel payment integration guide with no concealment, pretext, or covert behavior, but its default sandbox base URL uses plaintext http, which would send the merchant's x-api-secret unencrypted, a risk to the user's credentials under the recall-first override.

MEDIUM W009: Direct money access capability detected (payment gateways, crypto, banking).

  • Direct money access detected (medium risk: 0.30). The skill is a developer integration guide for Setel. Its core workflow covers building checkout sessions, verifying webhooks, and debugging, which the scope rules exclude as development guidance. However, the skill also documents operational payment routes that an agent could run with the merchant's credentials: creating checkout sessions for charges (line 28), capturing held payments (line 33), and refunds drawing on the merchant balance (line 50), with endpoints listed at line 42. Under the recall-first override, I treat these as a direct operation route. For direct operations, the service-enforced control is documented: partner authentication with x-api-key and x-api-secret is required on every request (line 39). That is a service-validated credential check, so it qualifies as an established control, which caps the tier at medium. The "Keys are backend-only" guidance is not itself the control; the service's credential validation is.

Issues (2)

E004
CRITICAL

Prompt injection detected in skill instructions.

W009
MEDIUM

Direct money access capability detected (payment gateways, crypto, banking).

Audit Metadata
Risk Level
CRITICAL
Analyzed
Nov 16, 2026, 05:13 AM
Issues
2