data-quality-auditor
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill's implementation is focused on statistical analysis and reporting. The Python scripts use standard libraries and perform no network requests, file system modifications, or privileged operations. The operations are limited to data parsing and mathematical profiling of dataset health.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted dataset files, which represents an indirect prompt injection surface.\n
- Ingestion points: Data is loaded from CSV files via the
--fileflag inscripts/data_profiler.py,scripts/missing_value_analyzer.py, andscripts/outlier_detector.py.\n - Boundary markers: Absent; there are no delimiters or instructions used to separate data content from the agent's core instructions.\n
- Capability inventory: Capabilities are restricted to statistical computation and standard output reporting; no network, file-write, or shell-execution capabilities exist within the tools.\n
- Sanitization: Absent; the scripts process input values as literal strings for type inference and statistical profiling.
Audit Metadata