The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests data from external websites and search results, which are untrusted sources that could contain malicious instructions designed to influence the agent. • Ingestion points: The scrape_url, search_engine, and extract_structured_data tools defined in SKILL.md. • Boundary markers: There are no explicit instructions or delimiters provided to help the agent distinguish between its instructions and the content being scraped. • Capability inventory: The agent processes the scraped HTML, Markdown, or JSON, potentially using it to drive further actions. • Sanitization: No sanitization or safety filtering of the scraped content is described in the skill configuration.
[EXTERNAL_DOWNLOADS]: The skill setup requires running @brightdata/mcp via npx, which downloads the package from the npm registry. This is an official resource from Bright Data, an established service provider.

brightdata-mcp