e2e-tester
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill requires the agent to navigate to target pages and take snapshots to identify selectors and user flows. This creates a surface where an attacker could embed malicious instructions in a web page to subvert the agent's task. (1) Ingestion points: Live DOM structure and page content processed via snapshots. (2) Boundary markers: The skill does not provide instructions for the agent to ignore or delimit instructions found within the UI content. (3) Capability inventory: The skill has access to Bash, Write, and Edit tools, which could be misused if the agent follows injected instructions. (4) Sanitization: No sanitization of UI labels or content is performed.
- [SAFE]: All provided code samples and instructions follow legitimate software testing best practices. There are no signs of obfuscation, hardcoded credentials, malicious persistence, or unauthorized data exfiltration. The use of command-line tools is limited to standard Playwright operations.
Audit Metadata