backend-nodejs
Pass
Audited by Gen Agent Trust Hub on Oct 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill triggers automatically based on the presence of local files like package.json or server.ts. This allows untrusted data from the project environment to enter the agent context. When combined with the skill's capabilities to write files and execute shell commands, it creates a surface where malicious project files could attempt to influence the agent's actions.
- Ingestion points: File existence checks for package.json, server.ts, index.ts, and src/server.ts.
- Boundary markers: No explicit boundaries or warnings are defined to distinguish project content from skill instructions.
- Capability inventory: The skill has access to Read, Write, Edit, Bash, Glob, and Grep tools.
- Sanitization: No specific sanitization or validation of the triggering files' content is defined.
- [EXTERNAL_DOWNLOADS]: The skill's metadata contains an installation command that retrieves the skill content from a public repository on GitHub, which is a well-known service.
- Evidence: The install_command in metadata.json uses curl to download content from raw.githubusercontent.com.
Audit Metadata