mercury-onboarding-apply
Fail
Audited by Snyk on Jun 13, 2026
Risk Level: HIGH
Full Analysis
HIGH W007: Insecure credential handling detected in skill instructions.
- Insecure credential handling detected (high risk: 1.00). The skill requires the LLM to write and submit user-supplied sensitive identifiers (full SSNs, EIN/Tax IDs, and other KYC PII) verbatim into apply.yaml and then pipe that file into
mercury apply, which forces the model to output secrets directly (high exfiltration risk).
MEDIUM W009: Direct money access capability detected (payment gateways, crypto, banking).
- Direct money access detected (high risk: 1.00). This skill is explicitly designed to interact with a bank onboarding flow: it uses the Mercury CLI to generate an apply.yaml, fills KYC/KYB data, and submits that file via
mercury applyto the Mercury API to start a real account onboarding (the API returns a signupLink). This is a direct integration with a banking service (onboarding/account-creation), not a generic toollike browser automation or a generic HTTP caller. Although it does not move funds, it performs bank account creation/submission — a banking API operation — and therefore constitutes direct financial execution authority.
Issues (2)
W007
HIGHInsecure credential handling detected in skill instructions.
W009
MEDIUMDirect money access capability detected (payment gateways, crypto, banking).
Audit Metadata