dependency-confusion-detect
Pass
Audited by Gen Agent Trust Hub on Aug 27, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructs the installation of external security tools from public sources. Specifically, it recommends
go install github.com/nickvdyck/confused@latestandpip install guarddog. These are recognized utilities in the security community for identifying supply chain vulnerabilities. - [COMMAND_EXECUTION]: The skill relies on shell command execution to perform its primary function. It uses the
confusedandguarddogCLIs to scan local files such aspackage.json,requirements.txt, andpom.xml. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data in the form of project dependency files.
- Ingestion points: Reads
package.json(npm),requirements.txt(Python), andpom.xml(Maven). - Boundary markers: None identified; files are passed directly to the scanning tools.
- Capability inventory: Executes shell commands and processes the output of security scanners.
- Sanitization: The skill relies on the underlying tools (
confused,guarddog) to safely parse the input files rather than interpreting the content as direct agent instructions.
Audit Metadata