token-integration-analyzer
Fail
Audited by Snyk on Jul 15, 2026
Risk Level: HIGH
Full Analysis
HIGH W007: Insecure credential handling detected in skill instructions.
- Insecure credential handling detected (high risk: 0.90). The skill explicitly asks for an RPC endpoint (often containing embedded API keys/secrets) and implies the agent will use it to perform on-chain queries, which would require the LLM to receive and potentially include that secret verbatim in requests or outputs.
Issues (1)
W007
HIGHInsecure credential handling detected in skill instructions.
Audit Metadata