skills/datadrivenconstruction/ddc_skills_for_ai_agents_in_construction/big-data-analysis/Gen Agent Trust Hub
big-data-analysis
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external project data in CSV and Parquet formats. This creates a surface for indirect prompt injection where malicious instructions could be embedded in the data records.\n
- Ingestion points: Data is loaded through the
load_from_parquetandstream_processmethods inSKILL.md.\n - Boundary markers: There are no explicit delimiters or instructions to the agent to ignore embedded commands within the data.\n
- Capability inventory: The skill has filesystem access to read and write files, and uses the pandas library for data manipulation.\n
- Sanitization: The skill does not perform validation or sanitization of the content within the data files beyond standard type conversion.\n- [COMMAND_EXECUTION]: The Python implementation uses pandas methods that interact with the local filesystem.\n
- Methods such as
pd.read_parquet,pd.read_csv, andpd.ExcelWriterallow reading from and writing to arbitrary file paths provided as arguments.
Audit Metadata