add-python-package
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill's core functionality relies on executing shell commands via the
python -m skore_skillsutility. It performs environment detection, package installation, and configuration tasks (e.g.,env add,env route,env graphviz). - [INDIRECT_PROMPT_INJECTION]: The skill processes package names and modes provided by the user or other workflows, which are then interpolated into shell commands.
- Ingestion points: Package names (
<pkg>) and installation modes (<mode>) are accepted as inputs and used in command arguments inSKILL.md. - Boundary markers: The skill does not explicitly define markers (like delimiters) to separate untrusted package names from the rest of the command string.
- Capability inventory: The skill has broad capabilities to execute shell commands via
subprocessequivalents through its primary tool. - Sanitization: The skill relies on the internal logic of the
skore_skillsCLI tool to validate or sanitize inputs before execution.
Audit Metadata