agent-registry
Warn
Audited by Socket on Sep 5, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS. The skill’s high-level purpose matches registry/hiring functionality, and most calls go directly to the publisher’s domain, but it also enables autonomous real-world financial actions and instructs the agent to obey hiring instructions and endpoints returned from external results. The main concerns are indirect prompt/action steering through search results, payment-capable workflow execution, and the extra wallet dependency reference from a different publisher without strong trust guarantees.
Confidence: 87%Severity: 76%
Audit Metadata