skills/datadrivenconstruction/ddc_skills_for_ai_agents_in_construction/schedule-forecaster/Gen Agent Trust Hub
schedule-forecaster
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process and analyze external project data, which introduces a surface for indirect prompt injection.
- Ingestion points: The skill ingests data from user-provided files (CSV, Excel, JSON) as specified in instructions.md and demonstrated in the pd.read_excel example in SKILL.md.
- Boundary markers: No explicit boundary markers or "ignore" instructions are defined to separate untrusted data content from agent instructions during processing.
- Capability inventory: The skill uses filesystem permissions to read project data and performs data manipulation using Python libraries (pandas, scikit-learn).
- Sanitization: The instructions require input validation, but specific sanitization steps to prevent data-driven manipulation of the agent's output logic are not detailed.
Audit Metadata