x-report-generator
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The script uses the Playwright library to automate a Chromium browser instance, allowing it to navigate web pages, scroll, and capture screenshots.
- [EXTERNAL_DOWNLOADS]: The skill requires the installation of the
playwrightpackage and its associated browser binaries viaplaywright install chromiumto function. - [INDIRECT_PROMPT_INJECTION]: The tool has an attack surface for indirect prompt injection as it ingests untrusted data from external social media posts.
- Ingestion points: Data is scraped from X (Twitter) search results and user profiles in
scripts/x_report_generator.py. - Boundary markers: Data is structured into
TweetDataobjects before processing. - Capability inventory: The skill can write files to the local disk (HTML and PNG reports) and execute JavaScript within the controlled browser context.
- Sanitization: The script uses Playwright's
inner_text()method which automatically strips HTML tags from the scraped content, reducing the risk of malicious code injection into the generated reports. - [DYNAMIC_EXECUTION]: The skill performs dynamic execution by injecting a JavaScript snippet into the browser's initial context to mask automation flags and by using
page.evaluate()to calculate layout dimensions and scroll the page. - [CREDENTIALS_UNSAFE]: The skill handles authentication by saving and loading browser cookies to a local file (
cookies.json). This is a standard practice for maintaining login state in browser automation and the script does not transmit these credentials to external servers.
Audit Metadata