You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
* Editorial: Voice and Tone - Changed "you" references to be more direct; Terminology - Corrected "MCP Servers" to "MCP servers"
* Editorial: Structural improvements
* Editorial: Voice and Tone (Use "you" for the reader), Structure (Default Page Structure 10/20/70 format) - Added intro line and restructured opening
* Update app/en/get-started/about-arcade/page.mdx
* Update landing page with marketing feedback and fix MDX errors
- Change hero title to "MCP Runtime for AI agents"
- Rename "Arcade Engine" to "Runtime" and "Authorized Tool Calling" to "Agent Authorization"
- Update Popular Integrations: remove Twitch, Spotify, Figma, Discord, Salesforce; add X, MS Teams, Outlook, HubSpot, Asana
- Add logo rows to QuickStartCards with side gradient fade effect
- Add gradient overlay to Popular Integrations section
- Remove sample app images (outdated logos)
- Center column titles, remove offset from Use Arcade column
- Fix MDX syntax errors in about-arcade and examples pages
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
* Editorial cleanup: update copy and fix formatting
- Remove editorial comment blocks from MDX files
- Fix indentation on tags props in example cards
- Update intro copy in examples and agent-frameworks pages
- Add newlines at end of files
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
---------
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Co-authored-by: RL "Nearest" Nabors <236306+nearestnabors@users.noreply.github.com>
Co-authored-by: Rachel Lee Nabors <nearestnabors@users.noreply.github.com>
Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>
description: "Learn how Arcade helps with auth and tool calling"
4
4
---
5
5
6
+
6
7
import { Tabs } from"nextra/components";
7
8
8
9
# About Arcade
9
10
10
-
Applications that use models to perform tasks (_agentic applications_) commonly require access to sensitive data and services. Authentication complexities often hinder AI from performing tasks that require user-specific information, like what emails you recently received or what's coming up on your calendar.
11
+
Applications that use models to perform tasks (_agentic applications_) commonly require access to sensitive data and services. Authentication complexities often hinder AI from performing tasks that require user-specific information, like what emails were recently received or what's coming up on a calendar.
11
12
12
-
To retrieve this information, agentic applications need to be able to authenticate and authorize access to external services you use like Gmail or Google Calendar.
13
+
To retrieve this information, agentic applications need to be able to authenticate and authorize access to external services like Gmail or Google Calendar.
13
14
14
-
Authenticating to retrieve information, however, is not the only challenge. Agentic applications also need to authenticate to **act** on your behalf - like sending an email or updating your calendar.
15
+
Authenticating to retrieve information, however, is not the only challenge. Agentic applications also need to authenticate to **act** on behalf of users - like sending an email or updating a calendar.
15
16
16
17
Without auth, AI agents are severely limited in what they can do.
17
18
18
19
## How Arcade solves this
19
20
20
-
Arcade provides an authorization system that handles OAuth 2.0, API keys, and user tokens needed by AI agents to access external services through tools. This means your AI agents can now act on behalf of users securely and privately.
21
+
Arcade provides an authorization system that handles OAuth 2.0, API keys, and user tokens needed by AI agents to access external services through tools. This means AI agents can now act on behalf of users securely and privately.
@@ -32,15 +33,15 @@ With Arcade, developers can now create agents that can _act as the end users of
32
33
33
34
## Auth permissions and scopes
34
35
35
-
Each tool in Arcade's MCP Servers has a set of required permissions - or, more commonly referred to in OAuth 2.0, **scopes**. For example, the [`Gmail.SendEmail`](/resources/integrations/productivity/gmail#gmailsendemail) tool requires the [`https://www.googleapis.com/auth/gmail.send`](https://developers.google.com/identity/protocols/oauth2/scopes#gmail) scope.
36
+
Each tool in Arcade's MCP servers has a set of required permissions - or, more commonly referred to in OAuth 2.0, **scopes**. For example, the [`Gmail.SendEmail`](/resources/integrations/productivity/gmail#gmailsendemail) tool requires the [`https://www.googleapis.com/auth/gmail.send`](https://developers.google.com/identity/protocols/oauth2/scopes#gmail) scope.
36
37
37
38
A scope is what the user has authorized someone else (in this case, the AI agent) to do on their behalf. In any OAuth 2.0-compatible service, each kind of action requires a different set of permissions. This gives the user fine-grained control over what data third-party services can access and what actions they can take in their accounts.
38
39
39
40
When an agent calls a tool, if the user has not granted the required permissions, Arcade Engine will automatically prompt the user to authorize the tool and coordinate the OAuth 2.0 flow with the service provider.
40
41
41
42
## How to implement OAuth 2.0-authorized tool calling
42
43
43
-
To learn how Arcade authorizes actions (tools) through OAuth 2.0 and how to implement your own auth flow, check out [Authorized Tool Calling](/guides/tool-calling/custom-apps/auth-tool-calling).
44
+
To learn how Arcade authorizes actions (tools) through OAuth 2.0 and how to implement auth flow, check out [Authorized Tool Calling](/guides/tool-calling/custom-apps/auth-tool-calling).
{/* Editorial: Structure - Added intro line explaining what readers will learn; Voice and tone - Removed marketing language "seamlessly" and "making it straightforward" */}
5
-
6
5
# Arcade with Agent Frameworks and MCP Clients
7
6
8
7
Arcade integrates with agent frameworks and MCP clients to add tool-calling capabilities to your AI applications.
9
8
9
+
These guides are for developers building AI applications who need to connect Arcade tools to their agent frameworks or MCP clients. These guides show you how to authenticate with Arcade, load tools, and execute them within your chosen framework. Each guide provides code examples and configuration steps to get you started quickly.
0 commit comments