code
Fail
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: HIGHCOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION] (HIGH): Shell injection vulnerability in
SKILL.md. The implementationcode ${ARGUMENTS:-.}interpolates user-controlled data directly into a shell environment. An attacker can use shell metacharacters (e.g.,;,&&, or backticks) to execute unauthorized commands. - [COMMAND_EXECUTION] (HIGH): Indirect Prompt Injection risk. The skill processes external input through
${ARGUMENTS}and has execution capabilities. Evidence: 1. Ingestion point:${ARGUMENTS}parameter. 2. Boundary markers: Absent. 3. Capability inventory: Arbitrary shell command execution via Bash. 4. Sanitization: None.
Recommendations
- AI detected serious security threats
Audit Metadata