NYC
skills/odyssey4me/agent-skills/gerrit/Gen Agent Trust Hub

gerrit

Fail

Audited by Gen Agent Trust Hub on Feb 15, 2026

Risk Level: HIGHEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [REMOTE_CODE_EXECUTION] (HIGH): The troubleshooting section advises downloading a script (commit-msg hook) from a remote placeholder URL (https://review.example.com/tools/hooks/commit-msg) and making it executable via chmod u+x. This pattern allows for the execution of arbitrary remote code.
  • [COMMAND_EXECUTION] (HIGH): Installation instructions for Debian/Ubuntu systems use sudo apt install, which requires operations with administrative privileges.
  • [PROMPT_INJECTION] (HIGH): The skill is vulnerable to Indirect Prompt Injection because it retrieves and processes commit messages and metadata from external Gerrit servers.
  • Ingestion points: Results from gerrit query and downloaded patchsets.
  • Boundary markers: No delimiters or explicit instructions to ignore embedded commands are present.
  • Capability inventory: The skill executes shell commands through git-review and ssh.
  • Sanitization: No input validation or sanitization is performed on data fetched from Gerrit.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Feb 15, 2026, 11:59 PM