arkit-spatial-sensing-workflow
Pass
Audited by Gen Agent Trust Hub on Oct 30, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill consists exclusively of markdown documentation and YAML metadata. It provides workflow instructions and coding standards but contains no executable scripts, binaries, or automated command execution.
- [DATA_EXFILTRATION]: The instructions include proactive safety guards that prohibit the persistence or transmission of sensitive spatial data, such as world maps and meshes, without explicit user consent and a clear purpose.
- [INDIRECT_PROMPT_INJECTION]: The skill defines a surface for ingesting untrusted data via the request input. However, the skill's purpose is to guide the agent in generating implementation plans and documentation rather than executing logic. There are no associated scripts or tools that would allow for command execution or malicious side effects from this input.
- [EXTERNAL_DOWNLOADS]: The documentation references well-known development tools and official Swift packages. These references are for developer guidance and do not trigger automated or unverified downloads by the skill itself.
Audit Metadata