The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the AI agent to use the terminal to execute the bdata (or brightdata) CLI for web searches, scraping operations, and managing account pipelines. This is the intended primary function of the skill for environments where MCP tools are unavailable. Evidence found in SKILL.md and references/cli-commands.md.
[EXTERNAL_DOWNLOADS]: The instructions describe using npx to execute vendor-provided packages (@brightdata/cli and @brightdata/mcp) from the NPM registry. These are recognized as legitimate resources from a well-known service provider. Evidence found in SKILL.md, references/cli-commands.md, and references/mcp-setup.md.
[PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection because it is specifically designed to fetch and process arbitrary data from external websites.
Ingestion points: Untrusted data enters the agent context via scrape_as_markdown, search_engine, and various bdata scrape/search commands as detailed in SKILL.md and references/cli-commands.md.
Boundary markers: The instructions do not specify the use of delimiters (such as XML tags or triple quotes) or explicit instructions to ignore embedded commands when processing retrieved web content.
Capability inventory: The skill has access to the terminal for shell command execution and performs network operations through the Bright Data API.
Sanitization: There are no instructions provided for sanitizing or validating external content before it is processed by the agent.

brightdata-guide