@6171/drift-detector
Pass
Audited by Gen Agent Trust Hub on Aug 8, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns or security risks were detected. The skill provides a legitimate utility for data reconciliation using standard Python libraries without external dependencies.
- [DATA_EXFILTRATION]: The skill was analyzed for data exfiltration; it reads local files specified by the user to perform comparisons but contains no network code, subprocess execution, or hardcoded credentials that would facilitate data theft.
- [PROMPT_INJECTION]: The skill was analyzed for indirect prompt injection risks. It ingests external data from JSON and CSV files, which could theoretically contain malicious instructions meant for the agent. However, the skill parses this data strictly as structured content and performs no dynamic execution. Ingestion points: 'old' and 'new' file arguments in 'scripts/drift.py'. Boundary markers: None. Capability inventory: Local file reading only (no command execution or network access). Sanitization: Uses standard library 'json' and 'csv' parsers.
Audit Metadata