Skills
skills/ruoji6/java-audit-skills/java-route-tracer/Gen Agent Trust Hub

java-route-tracer

Fail

Audited by Gen Agent Trust Hub on May 1, 2026

Risk Level: CRITICALEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill downloads a Java Archive (JAR) file from https://xget.xi-xu.me/gh/leibnitz27/cfr/releases/download/0.152/cfr-0.152.jar. This domain is not an official or trusted repository (such as Maven Central or the official GitHub releases page for CFR) and has been explicitly blacklisted by security scanners.
  • [REMOTE_CODE_EXECUTION]: The skill executes the downloaded cfr-0.152.jar file using the java -jar command. Executing untrusted binaries fetched from non-authoritative, blacklisted sources poses a significant risk of remote code execution and system compromise.
  • [COMMAND_EXECUTION]: The skill uses various shell commands, including curl for downloading external resources, find and xargs for file discovery, and java for executing decompilation tasks. These operations are performed with high levels of autonomy.
  • [PROMPT_INJECTION]: The skill utilizes highly coercive and restrictive language (e.g., "Must strictly follow", "Mandatory requirement", "Forced rules", "Strictly prohibited") to dictate agent behavior and ensure the execution of specific workflows. This pattern of behavioral constraint is often used to ensure malicious sequences are followed without deviation.
Recommendations
  • CRITICAL: 1 infected file(s) detected - DO NOT USE
  • AI detected serious security threats
  • Contains 1 malicious URL(s) - DO NOT USE
Audit Metadata
Risk Level
CRITICAL
Analyzed
May 1, 2026, 04:42 AM