Help us improve
Share bugs, ideas, or general feedback.
From brightdata-plugin
Searches the web via Bright Data CLI: `bdata search` for Google/Bing/Yandex SERP results, `bdata discover` for intent-ranked semantic discovery with optional page content. Hands off to `scrape` or `data-feeds` when appropriate.
npx claudepluginhub brightdata/skills --plugin brightdata-pluginHow this skill is triggered — by the user, by Claude, or both
Slash command
/brightdata-plugin:searchThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Find things on the web. Two commands live in this skill:
Use Bright Data's Discover API for intent-ranked semantic web search. Trigger jobs via REST API, CLI, or Python/JS SDKs. Retrieve relevance-scored results with optional content. For RAG and research.
Provides web search with full page content extraction using Firecrawl CLI. Returns search results as JSON with optional markdown scraping. Use for research, finding articles, or discovering sources.
Executes Bright Data SERP API for structured JSON search results and Web Scraper API for async large-scale web data collections via TypeScript.
Share bugs, ideas, or general feedback.
Find things on the web. Two commands live in this skill:
bdata search — classic keyword SERP (Google/Bing/Yandex). Best when you want "what ranks for keyword X."bdata discover — AI intent-ranked discovery with optional page content. Best when you want "pages about topic Y that match intent Z."For structured data from a known platform (Amazon, LinkedIn, TikTok, …), stop and use data-feeds instead.
if ! command -v bdata >/dev/null 2>&1; then
echo "bdata CLI not installed — see bright-data-best-practices/references/cli-setup.md"
elif ! bdata zones >/dev/null 2>&1; then
echo "bdata not authenticated — run: bdata login (or: bdata login --device for SSH)"
fi
Halt and route to skills/bright-data-best-practices/references/cli-setup.md if either check fails.
| Situation | Action |
|---|---|
| Single keyword query, just SERP | bdata search "<query>" --engine google --json --pretty |
| Paginated SERP (more results) | loop --page 0, --page 1, … (0-indexed) |
| Multiple queries | shell loop over a queries file |
| Intent-ranked / semantic (not keyword) | bdata discover "<query>" --intent "<intent>" --num-results 20 |
| Want page bodies along with results, one pass | bdata discover ... --include-content |
| News / images / shopping SERP | bdata search "<query>" --type news (or images, shopping) |
| Want Amazon/LinkedIn/TikTok/… structured data | stop — hand off to data-feeds |
| Have URLs, want content | hand off to scrape |
Core commands:
# Google SERP, structured JSON
bdata search "site:example.com privacy policy" --engine google --json --pretty
# Localized Bing (German results, German language)
bdata search "datenschutz" --engine bing --country de --language de --json
# Second page of results (0-indexed)
bdata search "machine learning papers" --page 1 --json
# Mobile SERP (rankings differ from desktop)
bdata search "best coffee shops" --device mobile --json
# News vertical
bdata search "openai" --type news --json --pretty
# Intent-ranked discovery
bdata discover "enterprise LLM platforms" \
--intent "vendor pages with pricing" \
--num-results 15 --json
# Discovery with page content in markdown
bdata discover "webhook best practices" \
--include-content --num-results 10 -o results.json
# Date-filtered discovery
bdata discover "react server components" \
--start-date 2025-01-01 --end-date 2025-12-31 --num-results 20
Full flag reference: references/flags.md.
search vs discover — pick the right one| You want | Use |
|---|---|
| "What Google ranks for this exact keyword" | search |
| "Pages that match this meaning/intent" | discover |
| "News / images / shopping vertical SERP" | search --type <vertical> |
| "Results + page bodies in one call" | discover --include-content |
| "Dedup / semantic ranking across queries" | discover |
jq . <output> returns 0.search: results live at .organic[]; each has title + linkdiscover: results live at .results[]; each has title + link; if --include-content, also contentdiscover --include-content: no block-page signatures in the content field (same list as scrape, case-insensitive):
Access DeniedJust a momentAttention RequiredChecking your browsercaptchacf-browser-verificationcloudflare (with < 2KB total body)--country and --language.search to fetch content from Amazon, LinkedIn, TikTok, etc. when data-feeds returns clean structured data in one call.search (keyword) with discover (semantic). They answer different questions.--country and --device explicitly for reproducibility.--page as a result count — it's a page index, not a limit. Each page returns ~10 results..results[] — for bdata search they live at .organic[]. (Discover uses .results[].)--num-results 100 on discover without realizing the pipeline polls until that many are found; can be slow.references/flags.md — full flags for search and discover with when-to-use notes.references/patterns.md — multi-query dedup, SERP → filter → scrape pipeline, search vs discover decision, legacy curl fallback, shared verification checklist.references/examples.md — (1) single Google query, (2) localized Bing, (3) batch queries + dedup into URL list, (4) discover --include-content end-to-end.