skills/datadrivenconstruction/ddc_skills_for_ai_agents_in_construction/cwicr-data-validator/Gen Agent Trust Hub
cwicr-data-validator
Pass
Audited by Gen Agent Trust Hub on Mar 5, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements data validation logic using the CWICRDataValidator class. It performs standard data integrity checks such as identifying missing required fields, negative values in numeric fields, and statistical outliers.
- [COMMAND_EXECUTION]: No shell commands or subprocess executions were detected. All operations are performed within the Python runtime using established data science libraries.
- [EXTERNAL_DOWNLOADS]: The skill does not perform any network operations. It processes data locally from files provided by the user or existing in the environment.
- [DATA_EXPOSURE]: The skill requires filesystem permissions to read datasets (Parquet, Excel) and write validation reports. This access is strictly limited to the intended functionality of data auditing.
- [INDIRECT_PROMPT_INJECTION]: The skill processes external data (Excel, CSV, JSON) which represents an ingestion surface. However, the data is processed as structured content via pandas, and there is no evidence of the data being used to dynamically construct system-level instructions or override agent behavior.
Audit Metadata