skills/ascend-ai-coding/awesome-ascend-skills/profiling-analysis-profiling-hostbound/Gen Agent Trust Hub
profiling-analysis-profiling-hostbound
Pass
Audited by Gen Agent Trust Hub on Apr 2, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns or security risks were identified in the skill content.
- [PROMPT_INJECTION]: The skill exhibits a potential surface for indirect prompt injection as it processes external CSV data via the
analyze_hostbound.pyscript. While no malicious intent is present, the following evidence chain is noted for risk assessment: - Ingestion points: The script
scripts/analyze_hostbound.pyaccepts CSV files via the--inputparameter. - Boundary markers: The provided documentation does not specify the use of delimiters or 'ignore embedded instructions' warnings for the data being analyzed.
- Capability inventory: The skill is designed to perform data analysis and generate reports on system performance.
- Sanitization: No information is provided regarding the validation or sanitization of the CSV content before processing.
Audit Metadata