skills/datadrivenconstruction/ddc_skills_for_ai_agents_in_construction/historical-cost-analyzer/Gen Agent Trust Hub
historical-cost-analyzer
Pass
Audited by Gen Agent Trust Hub on Mar 5, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements construction-specific cost analysis logic using standard Python libraries (pandas, numpy, and scipy). All logic is transparent, and no hidden or suspicious functionality was detected.
- [PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it is designed to ingest and analyze external construction project data (e.g., CSV or Excel files). This is an inherent trait of data processing skills.
- Ingestion points: Project names, locations, and type metadata loaded via the
load_datamethod andfind_similar_projectscriteria. - Boundary markers: The skill does not currently use specific delimiters or 'ignore instructions' warnings when processing project descriptions or outlier data.
- Capability inventory: The skill is limited to file system read access for data loading and statistical calculations. It lacks network or subprocess execution capabilities.
- Sanitization: The Python implementation performs data type normalization and statistical validation, though it does not explicitly filter natural language instructions within the project metadata.
Audit Metadata