synthesis-mac-sync

Pass

Audited by Gen Agent Trust Hub on Apr 15, 2026

Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to generate and execute comprehensive bash scripts to manage file synchronization, permissions (chmod 600), and Git operations. This includes a 'One-time actions' system that executes arbitrary shell commands provided in a configuration file.
  • [DATA_EXFILTRATION]: Configuration files, which are noted to potentially contain sensitive data such as API keys and shell profiles, are synchronized to iCloud Drive.
  • [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection by processing instructions and manifests from external, user-controllable files.
  • Ingestion points: The agent reads the synchronization manifest and 'One-time actions' list from a README.md and git-repos.yaml file located in the iCloud sync folder.
  • Boundary markers: No specific delimiters or instructions to ignore embedded commands are used when reading these external files.
  • Capability inventory: The skill leverages bash execution capabilities to perform file system writes, network-based Git operations (push/pull), and execution of arbitrary script blocks.
  • Sanitization: The skill does not define validation or sanitization routines for the content of the manifests or the bash commands found in the 'One-time actions' section before they are executed.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 15, 2026, 11:50 AM