ieee-journal-browse

Pass

Audited by Gen Agent Trust Hub on Mar 15, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface because it ingests untrusted data from a third-party website (IEEE Xplore).
  • Ingestion points: Data is collected from page headers, body text, and links via evaluate_script in SKILL.md.
  • Boundary markers: None. The scraped content (titles, metrics) is interpolated directly into the response template in Step 3.
  • Capability inventory: The skill uses navigate_page and evaluate_script tools.
  • Sanitization: No filtering or sanitization of the external web content is performed.
  • [COMMAND_EXECUTION]: The skill utilizes browser-side script execution to navigate and scrape pages.
  • Evidence: The initScript in SKILL.md contains code (Object.defineProperty(navigator, 'webdriver', {get: () => undefined})) designed to modify browser properties to evade anti-bot detection.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 15, 2026, 07:10 AM