code-execution

Warn

Audited by Gen Agent Trust Hub on Aug 21, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides an execution environment for Python scripts that interact directly with the local filesystem (fs), git repository (git), and code transformations (transform). This allows for arbitrary file modification and shell command execution via the git interface.
  • [EXTERNAL_DOWNLOADS]: The SKILL.md file contains instructions to run a setup script located at ~/.claude/plugins/marketplaces/mhattingpete-claude-skills/execution-runtime/setup.sh if the runtime is not installed. This references a specific user's repository (mhattingpete) on an external marketplace, which introduces a dependency on unverified third-party code that is executed with shell privileges.
  • [PRIVILEGE_ESCALATION]: While no explicit sudo commands are present, the installation of plugins and the ability to execute git commands and filesystem operations provides a broad attack surface for modifying system configurations or user environment files (e.g., .bashrc) if the agent is directed to do so through complex workflows.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Aug 21, 2026, 05:03 AM
Security Audit — agent-trust-hub — code-execution