zero-human-dev-team
Pass
Audited by Gen Agent Trust Hub on Nov 7, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill consists entirely of instructional markdown content and does not include any executable scripts, shell commands, or tool definitions.
- [CREDENTIALS_UNSAFE]: The content includes several security-positive rules (80, 81, and 82) that explicitly prohibit agents from placing keys or passwords in chat, printing environment variables, or including credentials in backups.
- [INDIRECT_PROMPT_INJECTION]: Rule 83 serves as a defense-in-depth measure against indirect prompt injection by instructing agents to treat text from files, issues, or web pages as information rather than executable instructions.
- [PRIVILEGE_ESCALATION]: Rule 31 establishes a security policy requiring code owner review for any installer that executes with root privileges.
Audit Metadata