manual-testing
Pass
Audited by Gen Agent Trust Hub on Oct 14, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill includes a set of Python scripts used for parsing and converting testing artifacts between various formats (Markdown, CSV, JSON, Excel, etc.). These scripts use
subprocess.call()to invoke other internal scripts using the current Python interpreter. This implementation uses argument lists rather than shell execution, which is a secure practice that prevents shell injection. - [INDIRECT_PROMPT_INJECTION]: The skill's primary function is to ingest and process project-specific context and requirements. This creates an attack surface for indirect prompt injection where a malicious requirement might attempt to influence agent behavior. However, the skill includes robust instructions for the agent to adhere to provided facts, mark assumptions clearly, and avoid inventing details, which serves as a significant defensive measure.
- [DYNAMIC_EXECUTION]: Runtime execution of the skill's internal script library occurs when format conversion or parsing is requested. This is used to transform data into user-specified formats using standard library modules and local helper files like
xlsx_reader.py. The path handling within the XLSX reader includes normalization checks to prevent path traversal issues (Zip Slip) during file processing.
Audit Metadata