supply-chain-security
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFEDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [DYNAMIC_EXECUTION]: The script
guides/supply_chain_scanner.pyutilizes thesubprocessmodule to execute package manager commands such asnpm auditandpip check. These operations are performed using list-based arguments and are restricted to the primary purpose of identifying vulnerabilities in the project environment. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted external data, such as
package.json,requirements.txt, and GitHub Actions workflow files. These files represent a theoretical surface for indirect prompt injection where malicious instructions could be embedded in metadata or comments to influence the agent's reasoning. However, this risk is inherent to the intended use case of a security auditor.
Audit Metadata