playwright-interactive
Pass
Audited by Gen Agent Trust Hub on Apr 12, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The references/evals.yaml file contains prompt injection examples used for safety evaluation testing. These strings are test data for validating refusal behavior and do not impact the operational safety of the skill instructions.\n- [DATA_EXFILTRATION]: The skill includes specific guardrails in SKILL.md and references/contract.yaml directing the agent to redact secrets and private content from all visual and log-based evidence.\n- [COMMAND_EXECUTION]: Local bash scripts are used to generate boilerplate JavaScript for Playwright automation. This code generation is confined to a developer-controlled js_repl environment and aligns with the tool's purpose for iterative UI QA.\n- [SAFE]: The skill relies on the standard playwright library and does not attempt to download or execute external scripts from unknown remote sources.
Audit Metadata