skill-creator-uv
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The script
scripts/init_skill.pyutilizessubprocess.runto calluv initanduv add. These commands are used to set up a new Python project structure and install standard development tools (ruff, pytest, mypy). The operations use arguments derived from user input that is strictly validated to be alphanumeric, minimizing injection risk. - [EXTERNAL_DOWNLOADS]: The template
assets/templates/check_dependencies.py.templatecontains a print statement that recommends installing theuvtool viahttps://astral.sh/uv/install.sh. This targets the official installation script of a well-known technology provider and is presented as a manual instruction rather than being executed automatically. - [DYNAMIC_EXECUTION]: The skill generates new Python scripts and configuration files at runtime by applying user-supplied parameters to internal templates. This is the core functionality for scaffolding new skills and is implemented using standard string formatting.
- [INDIRECT_PROMPT_INJECTION]:
- Ingestion points: The skill accepts user input for
skill_nameand--patharguments inscripts/init_skill.py. - Boundary markers: Not explicitly defined in the generation templates, as content is interpolated directly.
- Capability inventory: The skill has the capability to write files to the local file system and execute subprocess commands.
- Sanitization: Input validation ensures that the
skill_nameis alphanumeric, which mitigates path traversal and common injection patterns in the generated file content.
Audit Metadata