skills/datadrivenconstruction/ddc_skills_for_ai_agents_in_construction/ifc-data-extraction/Gen Agent Trust Hub
ifc-data-extraction
Pass
Audited by Gen Agent Trust Hub on Mar 5, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides functionality to export extracted BIM data to SQL databases using the
sqlalchemylibrary. This involves executing database commands based on a provided connection string, which is a standard but sensitive operation. - [DATA_EXFILTRATION]: The skill's primary purpose is to read data from IFC files and write it to various formats such as Excel, CSV, JSON, and SQL databases. This creates a functional pathway for data to be transferred from the local filesystem to other locations.
- [PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection because it parses untrusted IFC files containing user-defined strings (e.g., element names, descriptions, and property values).
- Ingestion points: Data is ingested from local IFC files via the
ifcopenshell.open()method inSKILL.md. - Boundary markers: No specific delimiters or instructions are used to tell the agent to ignore potentially malicious commands embedded in the BIM metadata.
- Capability inventory: The skill can write to the local filesystem and interact with external databases using
os,openpyxl, andsqlalchemyas shown inSKILL.md. - Sanitization: There is no evidence of sanitization or filtering of the text extracted from the IFC files before it is processed or exported.
Audit Metadata