asset-it-management
Pass
Audited by Gen Agent Trust Hub on Nov 6, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns or security risks were identified. The skill is designed to generate structural templates for asset tracking.
- [DATA_EXFILTRATION]: The skill includes explicit security notes to avoid filling in real names, salaries, medical, or banking data. It specifically instructs that linked accounts should store only IDs/relations and never passwords or tokens.
- [INDIRECT_PROMPT_INJECTION]: While the skill ingests user input to customize asset registers, it follows a structured approach that minimizes injection risks.
- Ingestion points: User responses to asset tracking questions (Step 2) and intent identification (Step 1).
- Boundary markers: The skill uses an internal YAML context to manage state and separates prose from machine-readable data outputs.
- Capability inventory: The skill is limited to generating text-based artifacts (CSV, SQL DDL, JSON Schema). It does not have access to file-write or network tools according to the tools list.
- Sanitization: The skill mandates the use of placeholders and synthetic data for examples.
- [COMMAND_EXECUTION]: No shell commands or script execution patterns were found. Artifacts generated (SQL DDL) are intended for manual user application rather than automated execution.
- [EXTERNAL_DOWNLOADS]: References to external repositories (e.g., github.com/sickn33) are for documentation and related skills within the same developer context and do not involve remote code execution.
Audit Metadata