writing-webdriverio-code
Pass
Audited by Gen Agent Trust Hub on Feb 20, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- PROMPT_INJECTION (LOW): Indirect Prompt Injection surface detected in the data processing workflow.
- Ingestion points: The skill instructions specify reading external files such as
.webdriverio-skills/project-context.md,.webdriverio-skills/custom-rules.md, and site analysis files inreferences/website-analysis/to determine test implementation details. - Boundary markers: Absent. The instructions do not define clear delimiters or specific warnings for the AI to ignore potential instructions embedded within these external context files.
- Capability inventory: The resulting output is executable JavaScript (WebdriverIO), which, when run by a test runner, has capabilities including network access (browser interactions) and potentially filesystem or environment access depending on the test configuration.
- Sanitization: Absent. There is no mention of sanitizing or validating the content of the project context files before they are used to generate code logic.
Audit Metadata