From cc-devkits
Search the web or scrape webpages via Serper using the plugin runtime script at skills/serper/scripts/run-serper.js.
How this skill is triggered — by the user, by Claude, or both
Slash command
/cc-devkits:serperThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
This skill follows a **plugin-first, script-first** runtime model.
This skill follows a plugin-first, script-first runtime model.
Canonical runtime entrypoint (Claude resolves paths relative to skill location):
node ./scripts/run-serper.js
Use this skill when you need:
search or scrape.--query--urlnode ./scripts/run-serper.js search \
--query "{{query}}" \
[--gl us] \
[--hl en] \
[--num 10] \
[--page 1] \
[--location "{{location}}"] \
[--timeout 30000]
Options:
--query (required): search query string--gl (optional): country code, default us--hl (optional): language code, default en--num (optional): number of results, default 10, max 100--page (optional): page number, default 1--location (optional): location context--timeout (optional): timeout in milliseconds, default 30000node ./scripts/run-serper.js scrape \
--url "{{url}}" \
[--markdown] \
[--timeout 30000]
Options:
--url (required): target URL--markdown (optional): request markdown content--timeout (optional): timeout in milliseconds, default 30000Set one of:
SERPER_API_KEY=your_api_key_here
# or
SERPER_API_KEYS=key1;key2;key3
Env resolution precedence:
process.env.claude/skills/serper/.env.claude/.env.env~/.claude/.envCommands return JSON to stdout.
Search response typically includes:
organic[]knowledgeGraphanswerBoxrelatedSearches[]Scrape response typically includes:
textmarkdownmetadatalinks[]images[]401: invalid API key429: rate limit exceedednode ./scripts/run-serper.js search --query "TypeScript best practices" --num 5
node ./scripts/run-serper.js scrape --url "https://example.com" --markdown
npx claudepluginhub tan-yong-sheng/cc-devkits --plugin cc-devkitsRuns SerpApi searches for Google, Bing, or YouTube as structured JSON. Python and Node.js examples for first searches, query testing, and result format learning.
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.
Provides web search, content extraction, crawling, and research using Tavily API. Use for retrieving current web info, URL content, or site crawling.