playwright-python
Pass
Audited by Gen Agent Trust Hub on Jul 26, 2026
Risk Level: SAFECREDENTIALS_UNSAFEEXTERNAL_DOWNLOADSDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [CREDENTIALS_UNSAFE]: The documentation contains illustrative hardcoded credentials ('hunter2') in code examples. While used as a placeholder, users should be reminded not to include real secrets in scripts.
- [EXTERNAL_DOWNLOADS]: The skill requires downloading browser binaries (Chromium, Firefox, WebKit) from official Playwright repositories during the setup process.
- [DATA_EXFILTRATION]: The skill scripts perform network requests to arbitrary URLs to capture screenshots, DOM structure, and accessibility snapshots. This is the primary function of the skill but involves accessing external resources.
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes untrusted data from the web.
- Ingestion points: The scripts
dom.py,screenshot.py, andsnapshot.pyretrieve content from external URLs provided by the user. - Boundary markers: No specific delimiters or instructions are provided to the agent to distinguish between its instructions and the retrieved web content.
- Capability inventory: The skill can capture pixels, HTML structure, and text content, which are then read by the agent.
- Sanitization: The skill does not sanitize or filter the content retrieved from remote web pages before it is presented to the agent.
Audit Metadata