layer2
Warn
Audited by Gen Agent Trust Hub on Feb 19, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTION
Full Analysis
- [Unverifiable Dependencies & Remote Code Execution] (MEDIUM): The skill contains an explicit instruction for the agent to fetch content from an untrusted external URL (https://ethskills.com/l2s/SKILL.md). Since the target is another skill definition file, this facilitates the execution of unvetted remote logic.
- [Dynamic Execution] (MEDIUM): By redirecting the agent to a remote resource for its core instructions, the skill allows for behavior that can be modified after static analysis, effectively bypassing local security controls.
Audit Metadata