python-pro

Fail

Audited by Gen Agent Trust Hub on Apr 7, 2026

Risk Level: HIGHEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Fetches the 'uv' package manager installer from the official Astral domain.- [REMOTE_CODE_EXECUTION]: Instructs the agent on using a piped shell command to install the 'uv' toolchain from its remote source.- [COMMAND_EXECUTION]: Provides recipes for the 'just' task runner and shell commands for 'uv' and 'mise' to manage development environments.- [PROMPT_INJECTION]: The skill processes user-supplied code and project configurations while maintaining shell access, creating a surface for indirect prompt injection.
  • Ingestion points: Python source code and project metadata files like 'pyproject.toml'.
  • Boundary markers: Absent. There are no specific instructions to use delimiters for untrusted data.
  • Capability inventory: Shell command execution via 'just' task recipes and the 'uv' toolchain.
  • Sanitization: Absent. Input data is not validated or sanitized before influencing the agent's shell operations.
Recommendations
  • HIGH: Downloads and executes remote code from: https://astral.sh/uv/install.sh - DO NOT USE without thorough review
Audit Metadata
Risk Level
HIGH
Analyzed
Apr 7, 2026, 03:51 PM