oxylabs-web-unblocker

Pass

Audited by Gen Agent Trust Hub on Feb 18, 2026

Risk Level: SAFEDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • Data Exposure & Exfiltration (LOW): The skill performs network operations to unblock.oxylabs.io and arbitrary user-defined URLs. While intended for its scraping purpose, this involves transmitting data through a third-party proxy.
  • Indirect Prompt Injection (LOW): The skill is highly susceptible to indirect prompt injection as its primary function is to ingest untrusted data from the web.
  • Ingestion points: Content retrieved via the proxy in examples.md (Python, Node.js, PHP, Go, Java implementations).
  • Boundary markers: Absent; the code examples do not provide delimiters or warnings to the LLM about the untrusted nature of the scraped content.
  • Capability inventory: Full network access via curl, requests, and fetch to retrieve any web content.
  • Sanitization: Absent; the skill lacks any logic to sanitize or filter the retrieved HTML/content before passing it to the agent.
  • Security Best Practices (LOW): All provided code examples (cURL, Python, Node.js, etc.) explicitly disable SSL/TLS certificate verification (e.g., -k, verify=False, NODE_TLS_REJECT_UNAUTHORIZED='0'). This is a significant security risk that facilitates Man-in-the-Middle (MitM) attacks, although it is documented as a requirement for the service's custom certificate handling.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 18, 2026, 07:53 PM