Gemini adds 14 Connected Apps for work, design and daily tasks

Google is expanding Gemini into a control surface for third-party software. According to Google’s announcement, 14 new Connected Apps are rolling out across productivity, creative and consumer services. The integrations let Gemini retrieve information and perform actions such as updating tasks, editing images and publishing websites.

After linking a provider in Gemini’s settings, users can invoke it in a chat with an @ mention. Available operations depend on the connector and the permissions granted to the linked account. Some connectors only retrieve information, while others can change external systems by creating documents, updating records or publishing content.

One chat, 14 services

Google’s published examples cover 13 of the 14 services; Zoho appears in the connector list without a sample prompt. The examples show the range of supported operations:

• Adobe: Adjust a photo’s lighting to resemble golden hour.
• Picsart: Generate a logo for an artisanal matcha café.
• Linear: List every high-priority issue assigned to the user.
• Airtable: Query open tasks in a specified base.
• monday.com: Create a board item and set its status.
• Webflow: Add a responsive FAQ section and publish the site.
• PandaDoc: Draft a services agreement with a specified contract value.
• Wispr Flow: Extract decisions from recent meetings.
• Experian: Check a credit rating and identify factors that changed during the month.
• Peloton: Find a 30-minute advanced strength class focused on core exercises.
• apartments.com: Search for two-bedroom rentals in Miami below $3,500.
• SeatGeek: Find local events with highly rated ticket deals.
• Squarespace: Brainstorm and search for domains for a new business.

MCP makes the batch possible

The connectors build on Google’s support for the Model Context Protocol, an open standard for connecting AI clients to external tools and data. An MCP server describes the tools it offers, the inputs they accept and the results they return. Gemini can use that shared structure to discover and call tools without requiring a unique interaction model for every provider.

Google highlighted MCP when it announced an earlier expansion involving Canva, OpenTable and Instacart. The protocol gives Google a repeatable way to add services as different as Linear and Peloton in the same release.

Providers still define authentication, permission scopes, input schemas, available actions and error handling. As a result, two MCP-based connectors can offer substantially different capabilities even when they use the same underlying protocol.

The catalog grows, with gaps

Gemini’s official directory remains smaller than those available for Claude and ChatGPT, which offer hundreds of MCP-based connections. Zapier MCP advertises access to more than 9,000 apps. Those figures cover integrations with different setup requirements, support models and action sets, so raw connector counts do not establish feature parity.

Before this rollout, Gemini’s third-party selection centered on services such as Dropbox, GitHub, Wix, Spotify, Instacart, OpenTable, Zillow and Viator. Adding Linear, Airtable, monday.com, PandaDoc and Wispr expands its coverage of routine workplace workflows. Slack, Notion and Salesforce remain absent from the built-in list described by Google and require a custom MCP connection.

Account rules may block workflows

Gemini’s built-in connectors authorize one account per provider. Linking a second account replaces the first, so anyone moving between personal and work accounts must disconnect and authorize the service again. That limitation can complicate workflows spanning multiple Airtable bases, Linear workspaces or other provider accounts.

Custom MCP servers can connect services outside Google’s official directory, but Google applies several restrictions:

• They require a personal Google Account.
• They are unavailable for work and school accounts.
• Gemini Apps Activity must be enabled.
• They currently support English only.
• They run inside Gemini Spark tasks rather than throughout the Gemini app.

Teams evaluating the integrations should verify which read and write actions each connector exposes, which account Gemini can authorize and whether the provider is available as a built-in app. Those constraints determine which workflows can run entirely from a Gemini chat and which still require custom integration work.