delivery-manager
Pass
Audited by Gen Agent Trust Hub on Nov 12, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides structured workflows for delivery management without any identified security risks. The core logic resides in local Python scripts that operate on provided data files.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests external data for metric calculation and risk assessment.
- Ingestion points:
scripts/delivery_metrics_tracker.py,scripts/dependency_mapper.py, andscripts/risk_register.pyread JSON data from file paths provided as arguments. - Boundary markers: Absent; the skill relies on standard JSON parsing.
- Capability inventory: The provided scripts are restricted to read-only file access and standard output. They contain no network operations, subprocess execution, or file-writing logic.
- Sanitization: No specific sanitization for prompt injection is present, but the lack of dangerous capabilities makes this surface low-risk.
- [EXTERNAL_DOWNLOADS]: The documentation references research papers and industry guidelines from well-known domains including
dora.devandcloud.google.com.
Audit Metadata