browser-extension-builder
Pass
Audited by Gen Agent Trust Hub on Oct 31, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill serves as a development framework for creating Chrome and Firefox extensions using Manifest V3. The provided code templates for background scripts, content scripts, and popups follow standard, documented extension APIs.
- [INDIRECT_PROMPT_INJECTION]: The skill provides templates for content scripts that ingest untrusted data from web pages (e.g.,
document.querySelector('.data')). - Ingestion points: Data is read from the DOM in
content.jsand passed to other extension components viachrome.runtime.sendMessage. - Boundary markers: No specific delimiters are used in the templates, but the skill is designed for developer use rather than direct runtime LLM integration.
- Capability inventory: The templates use
chrome.storagefor persistence,chrome.tabsfor navigation, andfetchfor network requests. - Sanitization: The skill contains explicit instructions to "never interpolate page/user text into innerHTML," demonstrating a security-conscious design for the generated code.
Audit Metadata