slide-archetype-gen
Pass
Audited by Gen Agent Trust Hub on Jun 13, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: No attempts to bypass safety filters or override agent constraints were found. The skill includes restrictive instructions to ensure the agent remains within the local archetype library for design references.
- [DATA_EXFILTRATION]: No sensitive file access or unauthorized network operations were detected. The skill contains a local shell script used to scan generated HTML files for external references to prevent accidental data leaks or unauthorized dependencies.
- [REMOTE_CODE_EXECUTION]: No remote script downloads or installations from untrusted sources are present. All design archetypes and implementation samples are bundled locally.
- [COMMAND_EXECUTION]: The provided utility script (
scripts/scan_real_info.sh) is a diagnostic tool meant for local linting of HTML files and does not perform any high-risk operations. - [EXTERNAL_DOWNLOADS]: The skill references well-known services like Google Fonts for typography, which are considered safe. No downloads from unknown or risky domains are present.
- [OBFUSCATION]: No obfuscated commands, Base64-encoded payloads, or hidden URLs were identified in the metadata or the archetype definitions.
Audit Metadata