trivy
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill documents the execution of the
trivyCLI tool to perform security audits on various targets including container images, local filesystems, and Kubernetes clusters. - [EXTERNAL_DOWNLOADS]: The instructions reference official project documentation and the standard GitHub Action provided by the vendor for CI/CD integration.
- [INDIRECT_PROMPT_INJECTION]: As a security scanning utility, the skill is designed to ingest and process data from untrusted external sources (such as public container images or third-party repositories). This is the primary intended purpose of the skill and follows standard security tool behavior.
- Ingestion points: Data ingestion occurs when scanning external resources via
trivy repo,trivy image, andtrivy fs(SKILL.md). - Boundary markers: No specific delimiters or boundary warnings are included in the provided shell command examples.
- Capability inventory: The skill utilizes the
trivybinary to perform analysis and output vulnerability reports. - Sanitization: Relies on the internal parsing and safety logic of the Trivy application itself when processing different file formats.
Audit Metadata