The flagged file is just a small drawing of a logo stored as text inside the app. It contains no web address and does nothing when the app runs. The scanner's warning did not hold up when the actual file contents were examined.
Technical evidence
Scanner reason: JavaScript analysis matched static JavaScript security signal javascript.xray.shady-link. The match applies to this repository.
Contextual assessment: The supplied source for this file, in both decoded and normalized representations, contains only an inline SVG graphic: a static vector path with a W3C namespace declaration and a fill attribute. There is no URL, fetch call, network primitive, obfuscated string, or executable behavior present in the provided content. The scanner's 'shady-link' signal did not retain matched literals, and nothing in the supplied evidence demonstrates any link or network access. This appears to be a self-hosted brand icon asset consistent with the project's stated fully-offline, no-data-leaves-the-device purpose.
This code asks the app's own local server for saved lorebook data and displays it in the editor. It never runs downloaded content as code. The scanner flagged it only because a network request and command-handling code happen to appear in the same file, and those commands are just predefined editor operations, not anything triggered by the downloaded data.
Technical evidence
Scanner reason: JavaScript analysis matched static JavaScript security signal javascript.download-to-execution. The match applies to this repository.
Contextual assessment: The flagged region contains a fetch call to a same-origin backend endpoint that retrieves world-info data as JSON, plus a second call that refreshes a settings list. The retrieved data is parsed as JSON, cached, and rendered into UI dropdown and editor views; there is no eval, Function constructor, script injection, or command execution sink consuming the retrieved data anywhere in the supplied source or expansions. The 'code execution sink' correlation appears to arise from slash-command registration objects in the same file, which are static command definitions with named callbacks and argument type declarations, not dynamic execution of fetched content. All network traffic targets the local SillyTavern backend with standard authenticated request headers, which matches the file's documented purpose of loading lorebook data for the World Editor.