wxt
Pass
Audited by Gen Agent Trust Hub on Oct 31, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill uses
npx wxt@latest initto scaffold new projects. This is a standard initialization procedure for the documented WXT framework, fetching the latest version from the official registry. - [DATA_EXFILTRATION]: The background service worker sends data to OpenAI's official API (
api.openai.com). This network operation is part of the primary functionality (AI PR Review) and targets a well-known service. - [INDIRECT_PROMPT_INJECTION]: The content script ingests data from GitHub web pages (
.diff-view) to be processed by an LLM. While this presents an indirect injection surface, the skill limits the data size and uses it for the specific, documented purpose of generating a review summary. - [CREDENTIALS_UNSAFE]: The code snippets demonstrate secure management of user-provided API keys, using extension storage (
storage.setItem) and password-type input fields in the UI rather than hardcoding secrets.
Audit Metadata