The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill's primary purpose is to ingest data from arbitrary external URLs and provide it to the agent as markdown, HTML, or screenshots, creating an attack surface for instructions hidden in web content.
Ingestion points: Content is retrieved from user-specified URLs via the bdata scrape CLI command or the api.brightdata.com/request REST endpoint.
Boundary markers: The skill does not provide or mandate any delimiters or explicit boundary instructions to help the agent differentiate between the fetched content and its own instructions.
Capability inventory: The skill utilizes CLI execution (bdata) and network requests to external domains (api.brightdata.com).
Sanitization: The skill explicitly states that "Nothing is parsed for you," indicating that raw, unsanitized content is passed directly to the agent's context.
[COMMAND_EXECUTION]: The skill instructs the agent to execute the bdata CLI tool, specifically using the bdata scrape command to perform web fetching operations.
[EXTERNAL_DOWNLOADS]: The skill makes network requests to api.brightdata.com. This domain is an official resource owned by the skill vendor, brightdata.

fetch