LIVE
Mistral Makes Large 4 Available as Open Weights29/10/26 · Mistral AI|Gemini 4 Carbon tops the Artificial Analysis Intelligence Index v529/10/26 · Google|Mistral Large 4 now available on Hugging Face29/10/26 · Mistral AI|Google Introduces "Gemini 4 Carbon" as Next Step in Its Model Lineup28/10/26 · Google|DeepMind unveils Genie 4, a new generation of world model27/10/26 · Google DeepMind|Anthropic files IPO prospectus, with risk disclosures taking center stage26/10/26 · Anthropic|Robot policy trained only in simulation matches top real-world systems26/10/26|OpenAI unveils GPT-6.1 Terra, a model built for computer use26/10/26 · OpenAI|New benchmark checks whether LLMs actually test their temporal medical claims29/10/26|An ODE-based sampling method speeds up discrete diffusion models29/10/26|OpenAI revises its Model Spec with five guiding values for AI behavior29/10/26 · OpenAI|Inside Proton's internal clash over its AI pivot29/10/26 · Proton|Mistral Makes Large 4 Available as Open Weights29/10/26 · Mistral AI|Gemini 4 Carbon tops the Artificial Analysis Intelligence Index v529/10/26 · Google|Mistral Large 4 now available on Hugging Face29/10/26 · Mistral AI|Google Introduces "Gemini 4 Carbon" as Next Step in Its Model Lineup28/10/26 · Google|DeepMind unveils Genie 4, a new generation of world model27/10/26 · Google DeepMind|Anthropic files IPO prospectus, with risk disclosures taking center stage26/10/26 · Anthropic|Robot policy trained only in simulation matches top real-world systems26/10/26|OpenAI unveils GPT-6.1 Terra, a model built for computer use26/10/26 · OpenAI|New benchmark checks whether LLMs actually test their temporal medical claims29/10/26|An ODE-based sampling method speeds up discrete diffusion models29/10/26|OpenAI revises its Model Spec with five guiding values for AI behavior29/10/26 · OpenAI|Inside Proton's internal clash over its AI pivot29/10/26 · Proton|
ToolRisingofficial publisher

Arcade.dev

Tool : A precise action the AI can call · requires arcade

7,500 tools across 81 MCP servers, with per-call OAuth and tokens never exposed to the model.

The best authentication model on the market.

GitHub · 1,058 ★ Docs
Overall score
8.4 /10
Security: one caveat, see the card
Utility
9.0
Reliability
10.0
Security
One caveat
Ease
5 min · Comfortable

Security reflects where data goes, requested permissions and incidents on record at the verification date. It is not a guarantee of zero risk: check the deployment mode and the connections you enable.

What it does, concretely

Connecting an AI agent to third-party services like Gmail, Slack, or GitHub requires managing OAuth tokens, storing them securely, and refreshing them without exposing them to the language model. Arcade.dev solves this by handling the entire authorization lifecycle on behalf of the developer.

In practice, each tool declares the OAuth scopes it needs through a simple Python decorator. Arcade then manages the authorization flow on the user side, stores the token in an encrypted environment, and injects it into the call at runtime. The language model and the client never see the token.

The platform offers more than 7,500 ready-to-use tools across 81 MCP servers, covering productivity, communication, databases, sales, and customer support. The open-source framework also allows developers to build their own MCP servers with internal integrations or custom OAuth providers.

Arcade.dev is aimed at Python developers building AI agents in production who need robust authorization management without handling it themselves. It is less suited to non-technical users or those looking for a no-code solution.

What it offers

  • ·Per-call OAuth management: scopes are declared per tool via a decorator, and the token is never passed to the model or the client.
  • ·Catalog of more than 7,500 ready-to-use tools across 81 MCP servers, covering Gmail, Slack, GitHub, Stripe, Notion, and many others.
  • ·22 built-in helper classes for common OAuth providers (GitHub, Google, Slack, Microsoft, Notion, LinkedIn, Zoom, etc.).
  • ·Open-source Python framework with a decorator API covering tools, resources, prompts, sampling, elicitation, progress, and logging.
  • ·arcade deploy command to host an MCP server on Arcade Cloud in a single step. (paid)
  • ·arcade evals command to test tool-call accuracy against real language models.
  • ·Compatibility with major agent frameworks: LangChain, CrewAI, OpenAI Agents, Google ADK, Mastra, Vercel AI SDK.
  • ·Centralized governance with shared registry, version control, visibility filtering, and OpenTelemetry audit logs. (paid)

In practice

Price
Freemium
Commercial use
Commercial use: yes
Available on
mcp, sdk-python
Host
arcade
FR interface
No

Security and data

Rating
Security: one caveat, see the card
Capabilities
readswritesexecutesnetwork
Your data
Data with third parties
Maintenance
Active

Links and repository checked on 25 October 2026 · added on 1 June 2024

How to install Arcade.dev

1. Make sure Python 3.10 or higher and uv are installed on your machine. 2. Install the Arcade CLI with: uv tool install arcade-mcp 3. Create a new MCP server with: arcade new my_server 4. Navigate to the generated folder and start the server in stdio mode with: uv run server.py 5. Configure your MCP client with: arcade configure claude (or cursor, or vscode depending on your tool). 6. Security note: never commit your .env file containing your keys; use the provided .env.example as a template.
uv tool install arcade-mcp

Official documentation ↗

In the same category

Whole category Automation & agents →