first-screen-check
Pass
Audited by Gen Agent Trust Hub on Nov 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes shell commands to perform technical website audits, including using
python3for HTML analysis,npx playwrightfor mobile viewport screenshot capture, andcurlfor verifying redirect chains. These operations are core to the skill's documented functionality. - [EXTERNAL_DOWNLOADS]: To enable browser automation, the skill invokes
npx playwright install chromium, which downloads browser binaries from official sources. This is a standard and expected behavior for rendering web content. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from user-provided URLs, which is a known attack surface for indirect prompt injection. * Ingestion points: Website HTML, text, and JSON-LD metadata are fetched via the
page_facts.pyscript and Playwright. * Boundary markers: The skill's output is structured into markdown files within the local directory. * Capability inventory: The skill can execute subprocesses (Python, Playwright, curl) and write to the local file system. * Sanitization: The skill relies on the underlying LLM's safety guardrails, as the provided scripts do not contain explicit content sanitization logic.
Audit Metadata