canhelp
Warn
Audited by Snyk on Nov 19, 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 workflow reads a mainnet canister's Candid metadata (
icp canister metadata ... candid:service), which anyone can publish by deploying a canister, but the agent only fetches it after the user supplies or selects a specific canister ID or name, so the attacker cannot force delivery.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata