artifact-proof
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: No security issues detected. The skill contains markdown instructions for software quality assurance, specifically focusing on verifying the integrity and functionality of build deliverables. No malicious patterns, command injections, or exfiltration vectors were identified.
- [INDIRECT_PROMPT_INJECTION]: The skill defines a workflow for processing and executing code from external artifacts. This represents a potential attack surface for indirect prompt injection, but as this is the primary intended function for artifact verification, it is considered safe in this context.
- Ingestion points: External build artifacts (ZIP, installers, executables) referenced in SKILL.md.
- Boundary markers: None explicitly defined in the procedural instructions.
- Capability inventory: Execution of artifact entrypoints and self-tests.
- Sanitization: None specified.
Audit Metadata