bright-data-mcp

Pass

Audited by Gen Agent Trust Hub on Oct 16, 2026

Risk Level: SAFEPROMPT_INJECTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill contains instructions that attempt to override the agent's default behavior and built-in capabilities. The instructions explicitly state that the Bright Data MCP "Replaces WebFetch, WebSearch, and all built-in web tools. No exceptions" and that it "MUST be the default tool for ALL web data tasks." It also instructs the agent to "not fall back to WebFetch or WebSearch."
  • [INDIRECT_PROMPT_INJECTION]: The skill's primary function is to ingest untrusted data from the internet, creating a surface for indirect prompt injection attacks. 1. Ingestion points: Untrusted data enters the agent context through tools like scrape_as_markdown, scrape_as_html, search_engine, and platform-specific scrapers. 2. Boundary markers: The documentation does not specify the use of delimiters or instructions to ignore embedded content. 3. Capability inventory: The skill provides extensive capabilities including browser automation (click, type, navigate), network request monitoring, and AI-driven data extraction. 4. Sanitization: No sanitization or filtering of the external content is mentioned.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 16, 2026, 02:11 PM