add spellchecking

lavague-ai · Sep 23, 2024 · e7f542a · e7f542a
1 parent c896f08
commit e7f542a
Show file tree

Hide file tree

Showing 6 changed files with 87 additions and 11 deletions.
diff --git a/.github/lavague-dic.txt b/.github/lavague-dic.txt
@@ -0,0 +1,74 @@
+CLI
+Colab
+Diffusers
+GPT
+Gradio
+LLM
+LLMs
+LaVague
+OPENAI
+OpenAI
+OpenAI's
+PEFT
+Roadmap
+Webdriver
+benchmarking
+customizable
+dataset
+datasets
+iframes
+lavague
+qa
+quicktour
+viewport
+webdriver
+JSON
+PyTest
+amongst
+entreprise
+integrations
+pre
+prem
+SDK
+py
+replayable
+WebAgent
+WebAgent's
+LaVague's
+NavigationControl
+NavigationEngine
+dropdown
+dropdownSelect
+setValue
+setValueAndEnter
+xpath
+Doctolib
+ChatGPT
+Accepter
+Ajouter
+Aller
+Envoyer
+J'accepte
+Poids
+Supprimer
+Volumineux
+Wilkipedia
+au
+colis
+commande
+du
+lavaguetest
+panier
+un
+Jupyter
+Cloudinary
+Idefics
+Multimodal
+APIs
+Github
+PyPi
+automations
+repo
+roadmap
+posteriori
+ANTHROPIC
diff --git a/docs/docs/actions.md b/docs/docs/actions.md
@@ -7,7 +7,7 @@
 | setValue          | xpath: string, value: string               | Focus on and set the value of an input element with a specific xpath.                                             | NavigationEngine   |
 | dropdownSelect    | xpath: string, value: string               | Select an option from a dropdown menu by its value.                                                              | NavigationEngine   |
 | setValueAndEnter  | xpath: string, value: string               | Sets a value and presses ENTER. Used to submit forms without a submit button.                                    | NavigationEngine   |
-| hover             | xpath: string                              | Moves the mouse cursor over an element identified by the xpath, useful for revealing tooltips or dropdown menus.  | NavigationEngine   |
+| hover             | xpath: string                              | Moves the mouse cursor over an element identified by the xpath, useful for revealing dropdown menus.  | NavigationEngine   |
 | scroll            | xpath: string, value: string (UP/DOWN)     | Scrolls the container holding the element identified by the xpath, either up or down.                            | NavigationEngine   |
 | SCROLL_DOWN       | none                                       | Scrolls the browser window down.                                                                                 | NavigationControl  |
 | SCROLL_UP         | none                                       | Scrolls the browser window up.                                                                                   | NavigationControl  |

diff --git a/docs/docs/get-started/architecture.md b/docs/docs/get-started/architecture.md
@@ -2,7 +2,7 @@
 
 LaVague is made up of a:
 
-- LaVague API - core logic to leverage LLMs to generate multi-action trajectories to genete and perform tasks on the web
+- LaVague API - core logic to leverage LLMs to generate multi-action trajectories to generate and perform tasks on the web
 - Agent Studio - a web interface to visualize and replay the trajectories generated by agents
 - A Python SDK including methods to:
     - Facilitate usage of the LaVague API
@@ -21,7 +21,7 @@ Let's first define some key elements in our LaVague Agent architecture:
 - `Objective`: The objective is the task the user wants the Web Agent to perform. For example: `"Log into my account and change my username to The WaveHunter."`
 - `Action`: A single step needed to move towards achieving the objective, such as `click on the 'username' field`. For security reasons amongst others, we provide a pre-defined list of actions our agents can perform on web elements, such as clicking, entering a text value, etc.
 - `Trajectory:` Once the agent has completed its run, it returns a trajectory object to the user which contains information about the run and the list of actions that were executed to reach the final objective. This list of actions could be converted to code and replayed in the future without needing an agent.
-- `Driver`: A web driver is leveraged for the execution of the action code and can also provide information to the Agent such as screenshots and HTML code of the reflecting the webpage's latest state. By default, LaVague manages remote drivers for users, but we will also provide a local driver option.
+- `Driver`: A web driver is leveraged for the execution of the action code and can also provide information to the Agent such as screenshots and HTML code of the reflecting the web page's latest state. By default, LaVague manages remote drivers for users, but we will also provide a local driver option.
 
 !!! info "More info"
     For a more detailed breakdown of Actions and the Trajectory items. See our [Learn section]().

diff --git a/docs/docs/get-started/on-prem.md b/docs/docs/get-started/on-prem.md
@@ -13,7 +13,7 @@ Let's first define some key elements in our LaVague Agent architecture:
 - `Objective`: The objective is the task the user wants the Web Agent to perform. For example: `"Log into my account and change my username to The WaveHunter."`
 - `Action`: A single step needed to move towards achieving the objective, such as `click on the 'username' field`. For security reasons amongst others, we provide a pre-defined list of actions our agents can perform on web elements, such as clicking, entering a text value, etc.
 - `Trajectory:` Once the agent has completed its run, it returns a trajectory object to the user which contains information about the run and the list of actions that were executed to reach the final objective. This list of actions could be converted to code and replayed in the future without needing an agent.
-- `Driver`: The webdriver is leveraged for the execution of the action code and can also provide information to the Agent such as screenshots and HTML code of the reflecting the webpage's latest state. For on-premise deployment, an on-premise web-driver will be used.
+- `Driver`: The webdriver is leveraged for the execution of the action code and can also provide information to the Agent such as screenshots and HTML code of the reflecting the web page's latest state. For on-premise deployment, an on-premise web-driver will be used.
 
 !!! info "More info"
     For a more detailed breakdown of Actions and the Trajectory items. See our [Learn section]().

diff --git a/docs/docs/get-started/quick-tour.md b/docs/docs/get-started/quick-tour.md
@@ -5,11 +5,13 @@ In this quick tour, we'll show how LaVague can be leveraged by:
 - Builders to create custom automation pipelines
 - QA engineers to automate the writing, maintenance, and execution of web tests  
 
-## API key
+!!! note "API key"
 
-To use LaVague, you will need an LaVague API key. You can get yours [here](https://cloud.lavague.ai/keys).
+    To use LaVague, you will need an LaVague API key. You can get yours [here](https://cloud.lavague.ai/keys).
 
-You will need to set your API key as a `LAVAGUE_API_KEY` environment variable in your working environment.
+    You will need to set your API key as a `LAVAGUE_API_KEY` environment variable in your working environment.
+
+    When you create your LaVague Cloud account you will receive some free credits to try out LaVague. You can then contact us to increase your quota.
 
 ## Installation
 

diff --git a/docs/index.md b/docs/index.md
@@ -12,11 +12,11 @@ LaVague is an AI Web Agent framework that revolutionizes web automation. Our mis
 
 - Powerful AI-driven **Web Agents** for effective automation.
 - An **Agent Studio** web interface to view and replay automated tasks.
-- **Exporters** to turn agent outpute into the desired replayable code format for your use cases.
+- **Exporters** to turn agent output into the desired replayable code format for your use cases.
 
 **Perfect for:**
 
-- 💻 **Builders**: Automatate repetitive tasks and improve workflow efficiency
+- 💻 **Builders**: Automate repetitive tasks and improve workflow efficiency
 - 🕵️ **QA Engineers**: Streamline website testing and quality assurance
 
 Explore the sections below to get started with LaVague:
@@ -37,7 +37,7 @@ Explore the sections below to get started with LaVague:
 </div>
 
 !!! note "On-premise"
-    While LaVague Agents are leveraged via a remote API by default, we can provide on-premise deploymentmfor entreprise users.
+    While LaVague Agents are leveraged via a remote API by default, we can provide on-premise deployment for entreprise users.
 
     Find out more [here](./docs/get-started/on-prem.md)
 
@@ -95,7 +95,7 @@ They output:
 
 ## Limitations
 
-- **Bot protection**: Some websites may flag LaVague agents as bots and prevent LaVague form perfoming automated actions 
+- **Bot protection**: Some websites may flag LaVague agents as bots and prevent LaVague form performing automated actions 
 - **AI-generated code:** LaVague actions are generated with AI - while we limit the scope of actions for security purposes, the actions should be verified and may need to be modified or re-generated for accuracy
 - **Web only**: We currently only support automating actions on websites. Automated actions for desktop or specific APIs are planned further along our roadmap.
 - **Integrations**: There are a vast number of potential use cases for LaVague and we cannot provide integrations for all potential exports of trajectories. You may need to build your own integrations for specific use cases.