setup

Warn

Audited by Gen Agent Trust Hub on Mar 15, 2026

Risk Level: MEDIUMCOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill modifies the ~/.claude/settings.json file to inject a complex multi-line shell script into the statusLine configuration. This script is designed to run automatically in the background, using tools like ls, uv, jq, and mcpl to gather and display environment metadata.
  • [COMMAND_EXECUTION]: Executes several shell commands for system preparation, including cat to read existing configurations, mkdir to create rules directories, and cp -r to copy template files from the plugin root to the user's home directory.
  • [COMMAND_EXECUTION]: Utilizes a Python one-liner (python3 -c) to programmatically parse and check for duplicate notification hooks within the user's settings file.
  • [COMMAND_EXECUTION]: Invokes a local Python script bundled with the skill (sync_context.py) using uv run to generate and update the assistant's core rule files.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 15, 2026, 01:56 AM