agent-native-project-design
Pass
Audited by Gen Agent Trust Hub on Sep 12, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides security-focused instructions for managing AI agents, including least-privilege principles for tool access and sanitization of untrusted inputs in CI environments.
- [SAFE]: Mentions of dynamic context injection (
!command) are purely educational, documenting platform features rather than attempting to execute unauthorized commands. - [SAFE]: All external URLs and project references point to trusted organizations (e.g., Anthropic, Vercel, Snyk) or established research repositories (e.g., arXiv, agentskills.io).
- [SAFE]: References to repositories under the 'efoo-team' namespace (e.g., 'l-shift', 'video-production-tabi') are consistent with the skill's declared author context and represent standard vendor functionality.
- [SAFE]: No obfuscation, data exfiltration patterns, or persistence mechanisms were found in any of the analyzed files.
Audit Metadata