Skip to content

Commit

Permalink
Documentation updates -- add introduction + Skyvern in action page (#477
Browse files Browse the repository at this point in the history
)
  • Loading branch information
suchintan committed Jun 16, 2024
1 parent 161e25b commit 6d6df2e
Show file tree
Hide file tree
Showing 4 changed files with 136 additions and 60 deletions.
5 changes: 3 additions & 2 deletions README.md
Original file line number Diff line number Diff line change
Expand Up @@ -25,20 +25,21 @@
<a href="https://www.linkedin.com/company/95726232"><img src="https://img.shields.io/badge/Follow%20 on%20LinkedIn-8A2BE2?logo=linkedin"/></a>
</p>

[Skyvern](https://www.skyvern.com) automates browser-based workflows using LLMs and computer vision. It provides a simple API endpoint to fully automate manual workflows, replacing brittle or unreliable automation solutions.
[Skyvern](https://www.skyvern.com) automates browser-based workflows using LLMs and computer vision. It provides a simple API endpoint to fully automate manual workflows on a large number of websites, replacing brittle or unreliable automation solutions.

<p align="center">
<img src="docs/images/geico_shu_recording_cropped.gif"/>
</p>

Traditional approaches to browser automations required writing custom scripts for websites, often relying on DOM parsing and XPath-based interactions which would break whenever the website layouts changed.

Instead of only relying on code-defined XPath interactions, Skyvern adds computer vision and LLMs to the mix to parse items in the viewport in real-time, create a plan for interaction and interact with them.
Instead of only relying on code-defined XPath interactions, Skyvern relies on prompts in addition to computer vision and LLMs to the mix to parse items in the viewport in real-time, create a plan for interaction and interact with them.

This approach gives us a few advantages:

1. Skyvern can operate on websites it’s never seen before, as it’s able to map visual elements to actions necessary to complete a workflow, without any customized code
1. Skyvern is resistant to website layout changes, as there are no pre-determined XPaths or other selectors our system is looking for while trying to navigate
1. Skyvern is able to take a single workflow and apply it to a large number of websites, as it’s able to reason through the interactions necessary to complete the workflow
1. Skyvern leverages LLMs to reason through interactions to ensure we can cover complex situations. Examples include:
1. If you wanted to get an auto insurance quote from Geico, the answer to a common question “Were you eligible to drive at 18?” could be inferred from the driver receiving their license at age 16
1. If you were doing competitor analysis, it’s understanding that an Arnold Palmer 22 oz can at 7/11 is almost definitely the same product as a 23 oz can at Gopuff (even though the sizes are slightly different, which could be a rounding error!)
Expand Down
49 changes: 49 additions & 0 deletions docs/getting-started/skyvern-in-action.mdx
Original file line number Diff line number Diff line change
@@ -0,0 +1,49 @@
---
title: 'Skyvern In Action'
description: 'Examples of Skyvern running in production'
---

## Automatically apply to jobs on sites like Lever.co
Skyvern can be instructed to navigate to job application websites like Lever.co and automatically generate answers, fill out and submit the job application.

[💡 See it for yourself](https://app.skyvern.com/create/job_application)

Video recording coming soon!

## Automate transactions on e-commerce websites
Skyvern can be instructed to navigate to e-commerce websites, search for products, and add them to cart. Skyvern can even purchase the underlying items!

[💡 See it for yourself](https://app.skyvern.com/create/finditparts)

<p align="center">
<img src="images/finditparts_recording_crop.gif" alt="Skyvern adding products to cart on finditparts.com" />
</p>

## Navigating to government websites to register accounts or fill out forms
Skyvern can navigate complex and boring government forms and fill them out with given information. It's especially powerful because the forms can change their layouts, and Skyvern will continue filling them out correctly.

[💡 See it for yourself](https://app.skyvern.com/create/california_edd)
```
./run_ui.sh california_edd
```
<p align="center">
<img src="images/edd_services.gif"/>
</p>

## Retrieving insurance quotes from insurance providers in any language
Skyvern can navigate complex multi-part forms and fill out forms in any language

[💡 See it for yourself](https://app.skyvern.com/create/geico)
<p align="center">
<img src="images/bci_seguros_recording.gif"/>
</p>

[💡 See it for yourself](https://app.skyvern.com/create/bci_seguros)
<p align="center">
<img src="images/geico_shu_recording_cropped.gif"/>
</p>

## Log into a portal and download invoices (Demo coming soon)
Skyvern can log into a website, navigate until it finds a page with invoices, and download the invoices

Demo video coming soon
104 changes: 60 additions & 44 deletions docs/introduction.mdx
Original file line number Diff line number Diff line change
@@ -1,71 +1,87 @@
---
title: Introduction
description: 'Welcome to the home of your new documentation'
description: '🐉 Automate Browser-based workflows using LLMs and Computer Vision 🐉'
---

<img
className="block dark:hidden"
src="/images/hero-light.svg"
alt="Hero Light"
/>
<img
className="hidden dark:block"
src="/images/hero-dark.svg"
alt="Hero Dark"
/>
[Skyvern](https://www.skyvern.com) automates browser-based workflows using LLMs and computer vision. It provides a simple API endpoint to fully automate manual workflows, replacing brittle or unreliable automation solutions.
<p align="center">
<img src="images/geico_shu_recording_cropped.gif" alt="Recording of Skyvern automating an insurance quote being generated on Geico.com" />
</p>

## Setting up
# Why Skyvern?

The first step to world-class documentation is setting up your editing environments.
Traditional approaches to browser automations required writing custom scripts for websites, often relying on DOM parsing and XPath-based interactions which would break whenever the website layouts changed.

Instead of only relying on code-defined XPath interactions, Skyvern relies on prompts in addition to computer vision and LLMs to the mix to parse items in the viewport in real-time, create a plan for interaction and interact with them.

This approach gives us a few advantages:

1. Skyvern can operate on websites it’s never seen before, as it’s able to map visual elements to actions necessary to complete a workflow, without any customized code
1. Skyvern is resistant to website layout changes, as there are no pre-determined XPaths or other selectors our system is looking for while trying to navigate
1. Skyvern is able to take a single workflow and apply it to a large number of websites, as it’s able to reason through the interactions necessary to complete the workflow
1. Skyvern leverages LLMs to reason through interactions to ensure we can cover complex situations. Examples include:
1. If you wanted to get an auto insurance quote from Geico, the answer to a common question “Were you eligible to drive at 18?” could be inferred from the driver receiving their license at age 16
1. If you were doing competitor analysis, it’s understanding that an Arnold Palmer 22 oz can at 7/11 is almost definitely the same product as a 23 oz can at Gopuff (even though the sizes are slightly different, which could be a rounding error!)

Want to see examples of Skyvern in action? [Check out some examples we have here](introduction/use-cases)

# How it works
Skyvern was inspired by the Task-Driven autonomous agent design popularized by [BabyAGI](https://github.com/yoheinakajima/babyagi) and [AutoGPT](https://github.com/Significant-Gravitas/AutoGPT) -- with one major bonus: we give Skyvern the ability to interact with websites using browser automation libraries like [Playwright](https://playwright.dev/).

<picture>
<img className="hidden dark:block" src="images/skyvern-system-diagram-dark.png" alt="Skyvern's system diagram"/>
<img className="block dark:hidden" src="images/skyvern-system-diagram-light.png" alt="Skyvern's system diagram" />
</picture>


## Skyvern use-cases

<CardGroup cols={2}>
<Card
title="Edit Your Docs"
icon="pen-to-square"
href="https://mintlify.com/docs/quickstart"
title="Automatically apply to jobs"
icon="magnifying-glass"
href="getting-started/skyvern-in-action#automatically-apply-to-jobs-on-sites-like-lever-co"
>
Get your docs set up locally for easy development
Watch Skyvern automatically apply to jobs
</Card>
<Card
title="Preview Changes"
icon="image"
href="https://mintlify.com/docs/development"
title="Automate e-commerce transactions"
icon="cart-shopping"
href="getting-started/skyvern-in-action#automate-transactions-on-e-commerce-websites"
>
Preview your changes before you push to make sure they're perfect
Watch Skyvern automate purchases on e-commerce websites
</Card>
</CardGroup>

## Make it yours

Update your docs to your brand and add valuable content for the best user conversion.

<CardGroup cols={2}>
<Card
title="Customize Style"
icon="palette"
href="https://mintlify.com/docs/settings/global"
title="Interact with government websites"
icon="landmark-dome"
href="getting-started/skyvern-in-action#navigating-to-government-websites-to-register-accounts-or-fill-out-forms"
>
Customize your docs to your company's colors and brands
Watch Skyvern automate interacting with government websites
</Card>
<Card
title="Reference APIs"
icon="code"
href="https://mintlify.com/docs/api-playground/openapi"
title="Generate insurance quotes"
icon="clipboard"
href="getting-started/skyvern-in-action#retrieving-insurance-quotes-from-insurance-providers-in-any-language"
>
Automatically generate endpoints from an OpenAPI spec
Watch Skyvern navigate complex multi-page forms in any language
</Card>
</CardGroup>

## Quickstart
<CardGroup cols={2}>
<Card
title="Add Components"
icon="screwdriver-wrench"
href="https://mintlify.com/docs/components/accordion"
title="Quickstart (Skyvern Cloud)"
icon="cloud"
href="getting-started/quickstart-skyvern-cloud"
>
Build interactive features and designs to guide your users
Start automating tasks with the hosted version of Skyvern
</Card>
<Card
title="Get Inspiration"
icon="stars"
href="https://mintlify.com/customers"
title="Quickstart (Open Source)"
icon="github"
href="getting-started/quickstart-skyvern-open-source"
>
Check out our showcase of our favorite documentation
Start automating tasks in your own cloud
</Card>
</CardGroup>

38 changes: 24 additions & 14 deletions docs/mint.json
Original file line number Diff line number Diff line change
Expand Up @@ -48,41 +48,51 @@
"name": "Skyvern Cloud",
"icon": "cloud",
"url": "https://app.skyvern.com"
},
{
"name": "Book a demo",
"icon": "calendar",
"url": "https://meetings.hubspot.com/suchintan"
}
],
"navigation": [
{
"group": "Get Started",
"group": "Getting Started",
"pages": [
"introduction",
"getting-started"
"getting-started/skyvern-in-action",
"getting-started/quickstart-skyvern-cloud",
"getting-started/quickstart-open-source",
"getting-started/supported-features"
]
},
{
"group": "Running Tasks",
"group": "Tasks",
"pages": [
"running-tasks/introduction",
"running-tasks/prompting-tips",
"running-tasks/advanced-features",
"running-tasks/api-spec",
"running-tasks/prompting-tips",
"running-tasks/faq"
"running-tasks/webhooks",
"running-tasks/visualizing-results",
"running-tasks/visualizing-results-advanced"
]
},
{
"group": "Visualizer",
"group": "Workflows (Beta)",
"pages": [
"visualizer/set-up",
"visualizer/tasks-and-steps",
"visualizer/debugging-a-step",
"visualizer/faq"
"workflows/introduction",
"workflows/creating-workflows",
"workflows/running-workflows",
"workflows/workflow-blocks"
]
},
{
"group": "Workflows",
"group": "Integrations",
"pages": [
"workflows/introduction",
"workflows/components",
"workflows/faq"
"integrations/zapier",
"integrations/make.com",
"integrations/api"
]
}
],
Expand Down

0 comments on commit 6d6df2e

Please sign in to comment.