skills/datadrivenconstruction/ddc_skills_for_ai_agents_in_construction/carbon-calculator/Gen Agent Trust Hub
carbon-calculator
Pass
Audited by Gen Agent Trust Hub on Mar 5, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: Technical review of the Python implementation in SKILL.md confirms it is a legitimate calculation utility. The code uses the pandas library for data handling and report generation, which is standard practice for this use case. No external network requests or unauthorized command executions are present.\n- [SAFE]: The skill has a surface for indirect prompt injection because it is designed to ingest data from user-provided files (CSV, Excel, JSON). However, the implementation is safe as it does not include high-risk capabilities like network exfiltration or shell access.\n
- Ingestion points: The instructions.md file specifies that the agent accepts project data and file paths for processing.\n
- Boundary markers: No specific delimiters are used, but the instructions focus on structured data processing.\n
- Capability inventory: Filesystem access is limited to writing reports via pandas.ExcelWriter as declared in claw.json.\n
- Sanitization: The logic uses defined enums (MaterialCategory) and structured data classes (MaterialInput), which provides a layer of validation against malformed inputs.
Audit Metadata