c4-container
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill serves as a static documentation template and instructional guide for the C4 architectural model. It contains no executable scripts, shell commands, or network-active components.
- [EXTERNAL_DOWNLOADS]: The skill includes links to c4model.com for reference purposes. This is a well-known industry resource for software architecture.
- [INDIRECT_PROMPT_INJECTION]: The skill represents a surface for indirect prompt injection as it instructs the agent to process deployment configurations like Dockerfiles and Kubernetes manifests. However, the risk is negligible as the skill is constrained to documentation generation and lacks dangerous capabilities (such as code execution, file system writes, or network requests) that could be exploited.
- Ingestion points: External component documentation files, Dockerfiles, and Kubernetes manifests (SKILL.md).
- Boundary markers: None explicitly defined for untrusted data.
- Capability inventory: No execution, network, or file-write capabilities identified.
- Sanitization: None identified, but unnecessary due to lack of exploitable capabilities.
Audit Metadata