Scavio Cookbook: 104 real-world examples for the Scavio search API

104 runnable examples -- AI agents, notebooks, no-code workflows, and a pure scraper for every endpoint -- that search Google, YouTube, Amazon, Walmart, Reddit, TikTok, and Instagram in real time, powered by the Scavio search API.

Built across every way you'd actually use the API: raw REST, the Python and JS/TS SDKs, plain requests scrapers, LangChain, the Vercel AI SDK, CrewAI, Agno, smolagents, Haystack, AutoGen, n8n, and MCP.

50 free API credits to start (one-time). No credit card required. Get your free key in 30 seconds

---

What's inside

• One API, seven platforms -- Google, YouTube, Amazon, Walmart, Reddit, TikTok, Instagram.
• Eleven integration surfaces -- pick the one that matches your stack.
• Free alternative to SerpAPI, ScraperAPI, Bright Data, Jungle Scout, Fakespot, GummySearch, Modash, Keepa, and $149/month SEO tools.

Surface	Folder	Examples	Lang
Raw REST (curl + jq)	recipes/rest	6	bash
Pure API scrapers (one per endpoint)	recipes/scrapers	32	Python
Python SDK (no LLM)	recipes/python-sdk	7	Python
JS/TS SDK (no LLM)	recipes/js-sdk	6	TypeScript
Vercel AI SDK	recipes/vercel-ai	4	TypeScript
LangChain agents	agents	14	Python
LangChain notebooks	notebooks	19	Jupyter
CrewAI	recipes/crewai	4	Python
Agno	recipes/agno	3	Python
smolagents	recipes/smolagents	2	Python
Haystack	recipes/haystack	2	Python
AutoGen	recipes/autogen	2	Python
n8n (no-code)	recipes/n8n	2	JSON
MCP	recipes/mcp	1	Python/JSON

Each recipes/ subfolder has its own README and its own requirements.txt / package.json -- so the frameworks never fight over dependencies.

Quick start

Pick the stack you already use. Every example reads SCAVIO_API_KEY from the environment (free key: https://dashboard.scavio.dev).

# Raw API -- no install
export SCAVIO_API_KEY="sk_..."
curl -s -X POST https://api.scavio.dev/api/v2/google \
  -H "Authorization: Bearer $SCAVIO_API_KEY" -H "Content-Type: application/json" \
  -d '{"query":"best running shoes 2026"}' | jq .

# Python SDK
pip install scavio
python -c "from scavio import ScavioClient; print(ScavioClient().google.search('hello')['organic_results'][0]['title'])"

# JS/TS SDK
npm install scavio
node -e "import('scavio').then(async({Scavio})=>console.log((await new Scavio().google.search({query:'hello'})).results[0].title))"

Then run a full example:

git clone https://github.com/scavio-ai/cookbooks.git && cd cookbooks
cp .env.example .env   # add SCAVIO_API_KEY (free) and OPENAI_API_KEY

# a LangChain agent
pip install -r requirements.txt
python agents/instagram-scout.py "specialty coffee creators, 10k-250k followers"

# a no-LLM SDK pipeline
pip install -r recipes/python-sdk/requirements.txt
python recipes/python-sdk/youtube_channel_dashboard.py "langchain agents tutorial"

Catalog

Raw REST -- recipes/rest (curl + jq, no LLM)

Example	Platform
google-serp-to-json.sh -- SERP to clean JSON (organic, PAA, related)	Google
amazon-price-snapshot.sh -- search to top-ASIN price/seller snapshot	Amazon
reddit-keyword-firehose.sh -- poll new posts, dedupe seen ids (cron)	Reddit
youtube-search-to-metadata.sh -- search to full video metadata	YouTube
instagram-profile-card.sh -- profile to scouting card	Instagram
tiktok-hashtag-to-csv.sh -- hashtag to CSV of top videos	TikTok

Python SDK -- recipes/python-sdk (no LLM)

Example	Platform
paa_tree_expander.py -- BFS related-search + PAA crawler to CSV	Google
amazon_bulk_catalog.py -- ASIN list to product CSV	Amazon
walmart_price_logger.py -- price snapshots to SQLite	Walmart
reddit_alert_watcher.py -- alert on new posts (JSON state)	Reddit
youtube_channel_dashboard.py -- a channel's top videos ranked by views	YouTube
instagram_follower_overlap.py -- accounts two profiles both follow	Instagram
tiktok_hashtag_trend_logger.py -- log hashtag reach to SQLite	TikTok

JS/TS SDK -- recipes/js-sdk (no LLM)

Example	Platform
google-news-feed.ts -- news search to JSON feed	Google
amazon-price-drop-webhook.ts -- lowest price to Discord/Slack	Amazon
reddit-to-slack.ts -- new threads to Slack Block Kit	Reddit
youtube-playlist-builder.ts -- search to ranked playlist	YouTube
instagram-engagement-rate.ts -- profile + posts to engagement rate	Instagram
tiktok-creator-scorecard.ts -- creator search to scorecard	TikTok

Vercel AI SDK -- recipes/vercel-ai

Example	Tools
research-stream.ts -- streamed, cited research answer	Google
reddit-sentiment-tool.ts -- brand/product sentiment	Reddit
instagram-influencer-vet.ts -- creator discovery + vetting	Instagram
shopping-route.ts -- Next.js shopping endpoint	Amazon + Walmart

LangChain -- agents (new) + notebooks (new)

New agent	Platform
instagram-scout.py -- creator discovery shortlist	Instagram
instagram-competitor-watch.py -- competitor content monitor	Instagram
local-deal-scout.py -- best Walmart deals in a category	Walmart
amazon-review-miner.py -- mine reviews for praise/complaints	Amazon
youtube-course-builder.py -- YouTube to ordered curriculum	YouTube
momentum-radar.py -- cross-platform momentum score	Multi

New notebook	Platform
instagram-hashtag-analyzer.ipynb -- rank hashtags by volume	Instagram
instagram-profile-analytics.ipynb -- engagement + theme snapshot	Instagram
amazon-bestseller-rank-tracker.ipynb -- category leaders	Amazon
google-knowledge-graph-extractor.ipynb -- structured entity facts	Google
tiktok-comment-sentiment.ipynb -- comment sentiment (SDK + LLM)	TikTok

(Plus the original 8 agents and 14 notebooks -- see their folders.)

Multi-agent frameworks

Example	Framework	Platforms
recipes/crewai/market-research-crew.py	CrewAI	Google + Reddit
recipes/crewai/product-launch-crew.py	CrewAI	Amazon + Walmart + YouTube
recipes/crewai/influencer-campaign-crew.py	CrewAI	TikTok + Instagram
recipes/crewai/seo-content-crew.py	CrewAI	Google
recipes/agno/agno-shopping-assistant.py	Agno	Amazon + Walmart
recipes/agno/agno-social-listener.py	Agno	Reddit + TikTok + Instagram
recipes/agno/agno-research-team.py	Agno	Google + YouTube
recipes/smolagents/smolagents-web-researcher.py	smolagents	Google
recipes/smolagents/smolagents-fact-checker.py	smolagents	Google
recipes/haystack/haystack-rag-websearch.ipynb	Haystack	Google
recipes/haystack/haystack-news-qa.py	Haystack	Google
recipes/autogen/autogen-shopping-groupchat.py	AutoGen	Amazon + Walmart + YouTube
recipes/autogen/autogen-trend-analyst.py	AutoGen	YouTube + Reddit

No-code + MCP

Example	What
recipes/n8n/reddit-mention-to-slack.json	Scheduled Reddit -> Slack workflow
recipes/n8n/amazon-price-drop-email.json	Scheduled Amazon price -> email workflow
recipes/mcp/mcp_python_client.py	Call the hosted Scavio MCP server from Python
recipes/mcp/mcp-config.example.json	Drop-in MCP config for Claude Desktop / Cursor

Using Scavio via MCP

Scavio runs a hosted MCP server at https://mcp.scavio.dev/mcp exposing all 33 endpoints as tools. Point any MCP client at it with your key in the x-api-key header -- see recipes/mcp.

Why Scavio vs. the alternatives

Need	Expensive tool	Scavio
SERP + People Also Ask	SerpAPI ($75/mo)	Free (50 calls)
Amazon product data	Jungle Scout ($49-129/mo)	One API call
YouTube video metadata	VidIQ/TubeBuddy ($7-49/mo)	One API call
Reddit lead / GTM tracking	GummySearch ($29-99/mo)	recipes/python-sdk/reddit_alert_watcher.py
TikTok / Instagram creator discovery	Modash ($99-399/mo)	recipes/crewai/influencer-campaign-crew.py
Price tracking	Keepa / CamelCamelCamel	recipes/python-sdk/walmart_price_logger.py
Brand monitoring	Brand24 ($99-299/mo)	recipes/vercel-ai/reddit-sentiment-tool.ts

SDK cheat sheet

# Python -- pip install scavio
from scavio import ScavioClient
c = ScavioClient()                       # reads SCAVIO_API_KEY
c.google.search("ai agents")
c.amazon.product("B0CHWRXH8B")           # ASIN
c.youtube.search("python tutorial")
c.reddit.search("serpapi alternative", sort="new")
c.instagram.profile(username="nike")
c.tiktok.search_users(keyword="coffee")

// JS/TS -- npm install scavio
import { Scavio } from "scavio";
const c = new Scavio();                   // reads SCAVIO_API_KEY
await c.google.search({ query: "ai agents" });
await c.amazon.product({ asin: "B0CHWRXH8B" });
await c.tiktok.searchUsers({ keyword: "coffee" });

FAQ

Is this a free alternative to SerpAPI or ScraperAPI? Yes. 50 free real-time credits to start (one-time, no card) across Google, Amazon, YouTube, Walmart, Reddit, TikTok, and Instagram -- enough to build and test a project. Paid plans add monthly credits when you need more.

Which frameworks are supported? LangChain, the Vercel AI SDK, CrewAI, Agno, smolagents, Haystack, and AutoGen all have first-class Scavio tools, plus a hosted MCP server and an n8n node. Every framework has at least one example here.

Does it work with OpenAI, Claude, Gemini, local models? Yes -- the LLM is yours to choose. Examples default to OpenAI; swap freely.

Is scraping Amazon / Instagram legal? You call Scavio's licensed real-time search API, not the platforms directly. Your IP never gets blocked.

Contributing

PRs welcome:

• One example per file; keep agents under ~200 lines, notebooks under ~15 cells.
• A docstring/markdown header with prerequisites and usage.
• Verified against the free tier. No emojis, no generated-by-AI attribution.

Resources

• Scavio Dashboard -- free API key
• Scavio Docs -- REST + MCP reference
• scavio on PyPI - scavio on npm

---

Keywords: ai search api, serpapi alternative free, scraperapi alternative, amazon product api python, walmart api, youtube data api alternative, reddit search api, tiktok api, instagram api, langchain search tool, vercel ai sdk tools, crewai tools, agno tools, smolagents web search, haystack websearch, autogen tools, n8n scavio node, mcp search server, people also ask api, google serp api, ai shopping agent, influencer discovery api, brand monitoring api, price tracking api, retail arbitrage, seo keyword research api.

Powered by Scavio -- the real-time search API for AI agents.