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

gerrit

Pass

Audited by Gen Agent Trust Hub on Mar 14, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill interacts with Gerrit via git-review and ssh. The wrapper script scripts/gerrit.py uses subprocess.run with argument lists to mitigate command injection.
  • [EXTERNAL_DOWNLOADS]: The skill instructs users to install git-review via pip and download commit-msg hooks from the Gerrit server. These are standard configuration steps.
  • [PROMPT_INJECTION]: The skill reads external data from code reviews, presenting a surface for indirect prompt injection. Ingestion: scripts/gerrit.py (Gerrit query results). Capability: subprocess.run (ssh). Sanitization: Absent. Boundary markers: Absent.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 14, 2026, 04:35 AM