skills/datadrivenconstruction/ddc_skills_for_ai_agents_in_construction/weather-impact-scheduler/Gen Agent Trust Hub
weather-impact-scheduler
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external construction project data, which introduces a potential surface for indirect prompt injection attacks.
- Ingestion points: The skill accepts data from users in CSV, Excel, and JSON formats, as well as direct text input, as specified in
instructions.md. - Boundary markers: No explicit delimiters or instructions to ignore embedded commands are defined in the agent's instructions to protect against malicious content within these data sources.
- Capability inventory: The skill possesses file system write capabilities via the
export_analysismethod inSKILL.md, which usespandas.to_excelto save reports to a user-definable path. - Sanitization: There is no evidence of data sanitization or validation logic in
SKILL.mdto ensure that input data does not contain instructions that could influence the agent's behavior at runtime.
Audit Metadata