No material or immediate-danger concern was identified in this review.
This advisory report describes what the named tools and review process found at one exact commit. Unknown or unobserved behavior may still exist.
0 immediate danger0 material4 low
What this review found
No material or immediate-danger item was identified.
Contextual expected matches (4)
Credential access and network transmission in one file
Expected behavior · high confidence
This is a documentation file describing how the extension works, not the working code itself. What it describes is normal: if you choose to use your own AI service, the extension saves the key you type in and sends it only to the service address you chose, for your own requests. The key is not sent anywhere hidden or to the developer, and it stays in your browser.
Technical evidence
Scanner reason: A credential source and an outbound network operation were detected in the same file.
Contextual assessment: The flagged file is technical documentation, not executable code, so no runtime data flow exists at this location. The described design is that a user-configured custom OpenAI-compatible backend sends the user's own API key as a Bearer credential only to the endpoint URL the user explicitly configured in the extension settings, and that key is kept in browser-local storage rather than in synced extension settings or chat metadata. The network destination and the credential source are both user-chosen and user-visible, the key is not routed to any hardcoded or third-party destination, and the default backend path goes through the host SillyTavern connection rather than a custom endpoint at all. This matches the documented purpose of an extension that performs its own LLM analysis calls and is the standard pattern for a bring-your-own-endpoint feature.
Credential access and network transmission in one file
Expected behavior · high confidence
The flagged item is a design document, not running code. It simply explains how the extension sends chat text to the AI service the user chose, which is the feature's advertised purpose.
Technical evidence
Scanner reason: A credential source and an outbound network operation were detected in the same file.
Contextual assessment: This candidate is a documentation file describing the re-scan feature's prompt schema and transport. The flagged line only documents that prompts are sent via the host's generation API or a direct fetch to the user-configured custom API. Documentation text contains no executable behavior, and the described transport matches the disclosed, user-configured backend implemented in the production script. No credential flow beyond the user-selected endpoint is described.
The scanner flagged that this file both makes network requests and can load code modules. In reality, the network activity is just loading decorative icons and calling the AI service the user configured, and the code loading only imports SillyTavern's own built-in scripts. Nothing downloaded from the internet is ever run as code.
Technical evidence
Scanner reason: JavaScript analysis matched static JavaScript security signal javascript.download-to-execution. The match applies to this repository.
Contextual assessment: The scanner correlates a network-retrieval primitive with a dynamic-execution sink in the same file. The supplied source shows the only network references are static image URLs for panel icons and the documented user-configured OpenAI-compatible API calls. The dynamic import() calls target SillyTavern's own bundled script and regex engine modules within the host application, not downloaded remote content. No retrieved bytes are ever converted into executable code, so there is no download-to-execution data flow.
Credential access and network transmission in one file
Expected behavior · high confidence
The extension stores your API key and uses it when talking to the AI service you personally set up in its settings. That is exactly what the feature is advertised to do, and the key is only ever sent to the service address you typed in — never to anyone else.
Technical evidence
Scanner reason: A credential source and an outbound network operation were detected in the same file.
Contextual assessment: The scanner flags an API key being read from extension secret storage and transmitted over the network in the same file. The supplied code shows the key is read only for the custom-backend feature and is sent exclusively as a Bearer Authorization header to the exact base URL the user themselves configured in the extension settings, for the documented purpose of powering the MBTI analysis calls. There is no secondary destination, no key logging, no key persistence to chat data, and no transmission to any endpoint other than the user-configured one. This matches the README's disclosed behavior and is proportionate to the stated purpose.