adobe-photoshop
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides instructional content for Adobe Photoshop automation without any malicious patterns detected. The metadata and descriptions align with the actual content.
- [DATA_EXPOSURE_AND_EXFILTRATION]: The skill includes multiple explicit directives across all files (SKILL.md, gemini.md, gpt.md) to never invent, copy, or handle Adobe account credentials, login tokens, or license strings. It does not contain any network operations or exfiltration patterns.
- [COMMAND_EXECUTION]: The code samples provided use standard Photoshop ExtendScript and UXP APIs for local file management (Folder.selectDialog, File.saveAs) and image processing, which is entirely consistent with the skill's stated purpose for design and image pipelines.
- [PROMPT_INJECTION]: No prompt injection or jailbreak patterns were identified. The instructional directives focus on operational hygiene, such as closing documents without saving changes and maintaining color profiles.
- [DYNAMIC_EXECUTION]: While the skill involves scripting (ExtendScript/UXP), it does not use unsafe execution patterns like eval() on untrusted external data or remote code loading.
Audit Metadata