> ## Documentation Index
> Fetch the complete documentation index at: https://praison.ai/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Context.dev

> Live web search, scraping, extraction, brand intelligence, monitors, and batches for PraisonAI agents

Give a PraisonAI agent current web data and structured website intelligence through Context.dev.

```python theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
from praisonaiagents import Agent
from praisonai_tools import ContextToolkit

agent = Agent(
    name="web-researcher",
    instructions="Use Context.dev for current web research and cite source URLs.",
    tools=ContextToolkit().get_tools(),
)

agent.start("Find Context.dev's latest public product updates and summarize them.")
```

```mermaid theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
graph LR
    User[User] --> Agent[PraisonAI Agent]
    Agent --> Tool[Context.dev Tool]
    Tool --> API[Live Web and Context API]
    API --> Result[Grounded Result]

    classDef input fill:#8B0000,stroke:#7C90A0,color:#fff
    classDef process fill:#189AB4,stroke:#7C90A0,color:#fff
    classDef output fill:#10B981,stroke:#7C90A0,color:#fff

    class User,Agent input
    class Tool,API process
    class Result output
```

## Quick Start

**1. Install the integration**

```bash theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
pip install "praisonai-tools[context]"
```

**2. Configure your API key**

```bash theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
export CONTEXT_DEV_API_KEY="your-context-api-key"
```

Create an API key in the [Context.dev dashboard](https://app.context.dev/).

**3. Add Context.dev to an agent**

```python theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
from praisonaiagents import Agent
from praisonai_tools import ContextSearch, ContextScrape

agent = Agent(
    name="researcher",
    instructions="Search first, then scrape the most useful sources.",
    tools=[ContextSearch(), ContextScrape()],
)

agent.start("What did Stripe launch most recently? Include source links.")
```

## Choose the Tools You Need

Use focused tools for small agents:

```python theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
from praisonai_tools import (
    ContextCrawl,
    ContextExtract,
    ContextParse,
    ContextScrape,
    ContextSearch,
    ContextSitemap,
)

tools = [
    ContextSearch(),
    ContextScrape(),
    ContextCrawl(),
    ContextSitemap(),
    ContextExtract(),
    ContextParse(),
]
```

Use a toolkit when an agent needs a broader capability set:

| Toolkit | Capabilities |
| - | - |
| `ContextWebToolkit` | Search, scraping, crawling, sitemap discovery, extraction, screenshots, and document parsing |
| `ContextBrandToolkit` | Brand profiles, style guides, fonts, NAICS, and SIC classification |
| `ContextMonitorToolkit` | Website monitors, runs, changes, and usage |
| `ContextBatchToolkit` | Large asynchronous jobs and result retrieval |
| `ContextToolkit` | The complete public Context.dev catalog |

The combined toolkit exposes 26 safe tools by default. Monitor and batch operations that change external state require an explicit opt-in:

```python theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
from praisonai_tools import ContextToolkit

tools = ContextToolkit(
    include_write_tools=True,
    allow_browser_actions=True,
).get_tools()
```

## Common Workflows

> [accordion] Search the live web
>
> ```python theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
> from praisonai_tools import ContextSearch
>
> result = ContextSearch().run(
>     query="latest official Stripe product announcements",
>     numResults=10,
>     includeDomains=["stripe.com"],
>     freshness="month",
> )
> ```

> [accordion] Extract structured data
>
> ```python theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
> from praisonai_tools import ContextExtract
>
> result = ContextExtract().run(
>     url="https://www.context.dev/pricing",
>     schema={
>         "type": "object",
>         "properties": {
>             "plans": {
>                 "type": "array",
>                 "items": {
>                     "type": "object",
>                     "properties": {
>                         "name": {"type": "string"},
>                         "price": {"type": "string"},
>                     },
>                     "required": ["name", "price"],
>                 },
>             }
>         },
>         "required": ["plans"],
>     },
> )
> ```

> [accordion] Parse a local document
>
> ```python theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
> from praisonai_tools import ContextParse
>
> result = ContextParse().run(
>     file_path="./report.pdf",
>     extension="pdf",
>     includeLinks=True,
> )
> ```

> [accordion] Process a large URL list
>
> ```python theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
> from praisonai_tools import ContextBatchToolkit
>
> batch_tools = ContextBatchToolkit(include_write_tools=True).get_tools()
> ```
>
> Pass `batch_tools` to an agent that needs to submit a job, check its status, and page through its results.

## Safety Defaults

* Browser actions are omitted from scrape tool schemas unless `allow_browser_actions=True`.
* Tools that create, update, run, cancel, or delete monitors and batches are omitted unless `include_write_tools=True`.
* API credentials are read from `CONTEXT_DEV_API_KEY`; `CONTEXT_API_KEY` remains supported as a legacy alias.
* Tool responses omit internal debug and request metadata before they reach the agent.

## Related Resources

> [card] [Context.dev API documentation](https://docs.context.dev)
>
> Explore endpoint inputs, outputs, and usage examples.

> [card] [Tools overview](https://praison.ai/docs/tools/tools.md)
>
> Learn how PraisonAI agents call tools.

> [card] [Firecrawl](https://praison.ai/docs/tools/external/firecrawl.md)
>
> Compare another web scraping integration.

> [card] [Exa](https://praison.ai/docs/tools/external/exa.md)
>
> Compare another web search integration.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.