minesweeper
Pass
Audited by Gen Agent Trust Hub on Oct 14, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The skill uses shell commands to run local Python scripts (minesweeper_data.py, download_report.py) and the pdftotext utility. User-supplied arguments like stock codes and search-derived URLs are interpolated directly into these commands.\n- [INDIRECT_PROMPT_INJECTION]: The skill ingests and analyzes data from external annual reports. This untrusted content is processed to determine risk signals, which constitutes an indirect prompt injection surface.\n
- Ingestion points: Fetches financial report PDFs from search results and processes their text content (SKILL.md Phase 2).\n
- Boundary markers: No explicit delimiters or instructions are provided to the agent to ignore potentially malicious content within the reports.\n
- Capability inventory: The skill possesses the ability to execute shell commands and write files to the local filesystem.\n
- Sanitization: No sanitization or validation of the extracted PDF text is mentioned before it is processed by the agent.\n- [EXTERNAL_DOWNLOADS]: The skill downloads financial documents from public disclosure platforms such as cninfo.com.cn and stockn.xueqiu.com. These are recognized financial services used to obtain the data required for the skill's operation.
Audit Metadata