bilingual-content

Pass

Audited by Gen Agent Trust Hub on Jun 24, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The workflow requires executing Python scripts (generate_bilingual_md.py and scrape_medium_article.py) using the uv runner to automate the scraping and generation process.
  • [EXTERNAL_DOWNLOADS]: The scrape_medium_article.py script uses the Playwright library to fetch content from Medium and download image files to the local resources/images/ directory.
  • [PROMPT_INJECTION]: The skill processes untrusted data from external websites which is subsequently used in translation workflows, creating a surface for indirect prompt injection.
  • Ingestion points: External article content fetched via Playwright in scripts/scrape_medium_article.py.
  • Boundary markers: None identified; the scraped content is interpolated directly into markdown templates without delimiters or warnings to ignore embedded instructions.
  • Capability inventory: The skill performs file system writes (json.dump, f.writelines, f.write) and network requests (browser-based scraping and image downloads).
  • Sanitization: No content filtering or escaping of instruction-like text is performed on the scraped content.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 24, 2026, 10:04 AM
Security Audit — agent-trust-hub — bilingual-content