legal-ai-audit-bundle
Pass
Audited by Gen Agent Trust Hub on Jul 14, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes a local Node.js script (
scripts/assemble-bundle.mjs) to automate the packaging of audit artifacts. It performs routine filesystem tasks such as directory creation, file copying, and writing metadata, which are necessary for its stated purpose of creating an audit bundle. - [DATA_EXFILTRATION]: No network activity or external egress was detected. The skill is explicitly configured for local data residency and includes documentation and logic to prevent the inclusion of sensitive pseudonymization maps in the audit logs, supporting data protection best practices.
- [EXTERNAL_DOWNLOADS]: The skill references external resources such as Microsoft's agent-governance-toolkit and other legal tech repositories in its documentation and roadmap sections for attribution and future planning. However, it does not perform any automated downloads or execution of remote scripts.
- [REMOTE_CODE_EXECUTION]: The logic is entirely contained within the provided local files. The skill does not use functions like
eval()orexec()on external data, nor does it fetch code from the internet to run at runtime.
Audit Metadata