python-executor

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the Bash tool to execute the belt CLI, which handles authentication via belt login and script execution via belt app run. These commands are standard for interacting with the inference.sh cloud service.
  • [EXTERNAL_DOWNLOADS]: The skill references installation instructions hosted on a GitHub repository (github.com/inference-sh). It also suggests installing a prerequisite skill using npx skills add belt-sh/cli. These external resources are official components of the platform being utilized.
  • [DYNAMIC_EXECUTION]: The core functionality of the skill is to run arbitrary Python code. To mitigate security risks, the execution is performed in a sandboxed, isolated subprocess on a remote server, as stated in the documentation.
  • [INDIRECT_PROMPT_INJECTION]: The skill accepts user-provided code through a defined JSON schema. While this presents an attack surface for code-based injections, the use of a remote sandbox environment restricts the potential impact to the isolated execution context.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 02:19 PM
Security Audit — agent-trust-hub — python-executor