nomadnet-pages
Pass
Audited by Gen Agent Trust Hub on Feb 18, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- Indirect Prompt Injection (LOW): The skill describes a platform where untrusted data from network requests is passed to scripts via environment variables (field_ and var_). The provided templates demonstrate using this data (e.g., os.environ.get) without sanitization, creating a potential attack surface. 1. Ingestion points: Environment variables in SKILL.md and references/page-structure.md. 2. Boundary markers: Absent in code templates. 3. Capability inventory: Subprocess calls (os.popen), database access (sqlite3), and file system operations (os.walk) in SKILL.md, references/authentication.md, and references/file-serving.md. 4. Sanitization: Absent; templates directly interpolate or echo environment variables.
- COMMAND_EXECUTION (SAFE): The skill includes instructions for making scripts executable (chmod +x) and provides shebang-based templates (Python/Bash). These are standard practices for the NomadNet platform and are intended for educational reference.
Audit Metadata