remote-control

Pass

Audited by Gen Agent Trust Hub on Aug 26, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses run.sh scripts to manage tmux sessions and execute claude remote-control commands. It interacts with a local projects registry (projects.json) to resolve paths and start sessions in specific directories.
  • [COMMAND_EXECUTION]: Provides functionality to run arbitrary bash commands in registered projects via the run_in_project MCP tool, which is a core feature of its remote-access design.
  • [SAFE]: Dependency management uses uv to install mcp and httpx from official registries during the installation phase, following standard development practices.
  • [SAFE]: The skill implements path resolution logic using Python scripts to handle project lookups, incorporating basic traversal checks for file reading operations.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 26, 2026, 06:00 PM
Security Audit — agent-trust-hub — remote-control