qwen-mm-plugins-omni-skill-creator
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFEDYNAMIC_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [DYNAMIC_EXECUTION]: The scripts
improve_description.pyandrun_eval.pyutilizeimportlib.utilto dynamically load the localutils.pymodule. This is performed for internal modularity within the skill's script directory rather than for executing untrusted external code. - [EXTERNAL_DOWNLOADS]: The skill makes network requests to OpenAI-compatible API endpoints (such as DashScope) to perform LLM-based tasks, including description optimization and trigger evaluation. These operations are core to the skill's stated purpose of automating skill creation.
- [COMMAND_EXECUTION]: The skill instructions direct the agent to execute several bundled Python scripts (
aggregate_benchmark.py,package_skill.py,run_loop.py) for managing the skill lifecycle, zipping assets, and running evaluations. These scripts are contained within the skill's own package and operate on local data.
Audit Metadata