skills/datadrivenconstruction/ddc_skills_for_ai_agents_in_construction/historical-data-manager/Gen Agent Trust Hub
historical-data-manager
Pass
Audited by Gen Agent Trust Hub on Mar 5, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection through the processing of external legacy data.
- Ingestion points: Data is ingested from external file formats (Excel, CSV, XER, Access) via the
extract_from_legacy_excel,extract_from_csv, andread_primavera_xermethods inSKILL.md. - Boundary markers: The implementation lacks technical delimiters or explicit boundary markers to prevent the agent from interpreting instructions embedded within the historical records.
- Capability inventory: The skill possesses filesystem read permissions but does not contain capabilities for network operations, subprocess execution, or dynamic code evaluation.
- Sanitization: No sanitization or filtering is applied to the extracted project data (such as project names or notes) before it is structured into records or migration reports.
Audit Metadata