Agno Integration

Scavio ships as a native toolkit inside Agno, the high-performance Python framework for building agents. Import ScavioTools and hand it to any Agno Agent to give it real-time search across Google, YouTube, Amazon, Walmart, Reddit, TikTok, TikTok Shop, Instagram, X, and LinkedIn — a cost-effective Tavily and SerpAPI alternative, with one toolkit, one API key, and no custom HTTP code.

Introduction

The ScavioTools toolkit lives in the agno package itself and calls the official scavio Python SDK under the hood, so there is no HTTP code to write. You need Python 3.9 or later and a Scavio API key from dashboard.scavio.dev.

Every provider is enabled by default, and each is gated by an enable_* flag, so you can expose exactly the tools you want to the model. Hand the toolkit to an Agent and it decides which providers to call, runs the searches, and reasons over the combined results in a single run.

Step-by-Step Integration Guide

Step 1: Install

pip install agno scavio

The toolkit lives in the agno package; the scavio package is the official Python SDK it calls under the hood. Requires Python 3.9 or later.

Step 2: Set your API key

Get a key at dashboard.scavio.dev (free credits, no card), then set it as an environment variable:

export SCAVIO_API_KEY=sk_live_your_key

ScavioTools reads SCAVIO_API_KEY from the environment. You can also pass it explicitly: ScavioTools(api_key="sk_live_...").

Step 3: Basic usage

from agno.agent import Agent
from agno.models.openai import OpenAIChat
from agno.tools.scavio import ScavioTools

agent = Agent(
    model=OpenAIChat(id="gpt-5.5"),
    tools=[ScavioTools()],
    markdown=True,
)

agent.print_response("Find the GitHub repo for the Agno framework and summarize it")

Available Tools

Each Scavio endpoint is exposed as a provider-prefixed tool. Every provider is enabled by default; each is gated by an enable_* flag, so you can register a lean tool list:

# Web-only agent: Google, YouTube, Reddit
agent = Agent(
    tools=[
        ScavioTools(
            enable_google=True,
            enable_youtube=True,
            enable_reddit=True,
            enable_amazon=False,
            enable_walmart=False,
            enable_tiktok=False,
            enable_instagram=False,
        )
    ],
)

# Or register every tool explicitly
agent = Agent(tools=[ScavioTools(all=True)])

Tool names match the method names in the table below.

The model fills in each tool's arguments from the conversation; you rarely call these directly. Knowing the shape helps when you write instructions or debug a run:

Advanced Example

Give the toolkit to an agent with a model and a few instructions, and it will decide which providers to call, run the searches, and reason over the combined results in a single run:

from agno.agent import Agent
from agno.models.openai import OpenAIChat
from agno.tools.scavio import ScavioTools

agent = Agent(
    model=OpenAIChat(id="gpt-5.5"),
    tools=[ScavioTools()],
    instructions=[
        "Search the web and Reddit before you answer.",
        "Always cite the sources you used.",
    ],
    markdown=True,
)

agent.print_response(
    "Is the Agno framework worth using in 2026? "
    "Summarize what the docs claim and what developers on Reddit actually say."
)

Here the agent calls google_search and reddit_search on its own, then synthesizes one answer. Swap in a commerce question and it reaches for amazon_search and walmart_search instead — you do not wire up any of that routing yourself.

Works with any model

ScavioTools is model-agnostic. Anywhere Agno runs — OpenAI, Anthropic, Gemini, Groq, Ollama, or any other supported provider — the toolkit works unchanged. Point the Agent at a different model and the same tools come along.

from agno.models.anthropic import Claude

agent = Agent(model=Claude(id="claude-sonnet-4-5"), tools=[ScavioTools()])

How it works

Every tool returns the Scavio response as a JSON string the model can read directly. Errors are caught and returned as {"error": "..."} rather than raised, so a failed call never crashes the agent run. Calls go through the scavio SDK, which handles auth, rate limiting, and request formatting.

Credit costs

Most calls cost 1 credit, including every Google search. Instagram is priced per endpoint: 10 credits for most calls, 8 for post details and comment replies, and 2 for user posts. See the rate limits reference for plan limits and the errors reference for retry guidance.

Benefits of Scavio + Agno

• Native toolkit: import ScavioTools and hand it to any Agent — no custom HTTP code.
• Seven platforms, one key: Google, YouTube, Amazon, Walmart, Reddit, TikTok, and Instagram behind a single API key.
• Model-agnostic: works unchanged with OpenAI, Anthropic, Gemini, Groq, Ollama, and every other Agno-supported provider.
• Cost-effective: most calls cost a single credit — a Tavily and SerpAPI alternative with broader platform coverage.

Next Steps

• Scavio API quickstart — keys, credits, and your first request
• Google Search API reference — the endpoint behind google_search
• MCP Integration — every Scavio endpoint as a tool
• Agno documentation
• Agno on GitHub
• scavio SDK on PyPI