browserstack
Pass
Audited by Gen Agent Trust Hub on Sep 12, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes standard development commands including
npx playwright testandnpm install -D browserstack-localto manage testing and dependencies. - [INDIRECT_PROMPT_INJECTION]: The skill processes external data retrieved from BrowserStack's APIs, which could theoretically contain instructions if the application under test generates malicious log output.
- Ingestion points: Metadata and logs retrieved via tools such as
browserstack_get_logsandbrowserstack_get_sessionsinSKILL.md. - Boundary markers: None explicitly defined in the instructions for delimiting external tool output.
- Capability inventory: The agent is authorized to modify local configuration files (
playwright.config.ts) and execute shell commands (npm,npx). - Sanitization: The skill does not describe specific filtering or sanitization of the logs retrieved from the BrowserStack service.
Audit Metadata