skills/datadrivenconstruction/ddc_skills_for_ai_agents_in_construction/schedule-compression/Gen Agent Trust Hub
schedule-compression
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill provides a robust mathematical implementation for Critical Path Method (CPM) schedule compression. The code in
SKILL.mduses standard Python libraries for data structures and contains no network operations, credential exposures, or suspicious command executions.- [INDIRECT_PROMPT_INJECTION]: The skill allows for the ingestion of user-provided project data in formats such as CSV, Excel, and JSON, which constitutes a surface for indirect prompt injection. This is a common and necessary feature for its stated purpose. - Ingestion points: Data is accepted via file paths or direct input in CSV, Excel, and JSON formats as noted in
instructions.md. - Boundary markers: Not explicitly defined in the prompt templates.
- Capability inventory: The skill is granted
filesystempermissions inclaw.json, allowing it to read project files. - Sanitization: The instructions specify that the agent should validate inputs before processing.
Audit Metadata