The Agent Skills Directory

[SAFE]: The skill facilitates legitimate operations within Adobe Photoshop, such as creating, updating, and querying text layer properties. All operations are confined to the Photoshop environment via vendor-provided modules like adobe.photoshop and dcc_mcp_core.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted strings via the content parameter to populate Photoshop layers. However, the skill does not possess any high-privilege capabilities such as arbitrary command execution, file system modification, or network access that could be exploited by such content. The attack surface is considered safe as the data flow is restricted to UI elements within Photoshop.

photoshop-text