Terms

Model Context Protocol

Follow the latest coverage, related explainers and connected technology stories.

Latest coverage

CN
How do companies track the appearance of their software tools in AI recommendations?

How do companies track the appearance of their software tools in AI recommendations?

As developers increasingly rely on AI agents to search for tools and content management platforms, a practical case study by Webflow shows how to monitor and track the appearance rates, descriptions, and source accuracy of its tools across various models such as Claude, ChatGPT, and Perplexity.

CN
Startup Mintive Raises $7 Million and Launches Identity Management Platform for AI Agents

Startup Mintive Raises $7 Million and Launches Identity Management Platform for AI Agents

Mintive, a startup specializing in identity and security management for AI agents, has successfully raised a total of $11 million, backed by prominent investors such as Cowboy Ventures and experts like Jeff Dean and Andrej Karpathy.

CN
Canva Launches "Canva Craft": Its First Proprietary AI Model for High-Precision Image Generation

Canva Launches "Canva Craft": Its First Proprietary AI Model for High-Precision Image Generation

Global design platform Canva has announced the launch of its proprietary image generation model, "Canva Craft," which relies on a two-stage architecture that allows for design planning before execution to ensure text accuracy and element distribution.

CN
Key Cybersecurity Developments: Attack on Industrial Systems and Vulnerabilities in XWorm, Chrome, and Langflow

Key Cybersecurity Developments: Attack on Industrial Systems and Vulnerabilities in XWorm, Chrome, and Langflow

A SecurityWeek summary reviews a series of security developments, including an ICS-Sniper attack on SCADA system traffic, a remote command execution vulnerability in the XWorm control panel, and a Chrome browser exploit chain that earned a $250,000 bounty. The developments also included active exploitation of a Langflow vulnerability, data breaches, and campaigns against scam centers in West Africa.

CN
Workday Opens HR and Financial Data to External AI Agents

Workday Opens HR and Financial Data to External AI Agents

Workday announced the general availability of an MCP server, a SQL interface, and a shared agent runtime, enabling the creation of HR and finance agents within tools such as Claude Code and Cursor while enforcing user permissions and maintaining approval and audit records.

CN
Pluto Security Raises $42 Million to Secure Companies’ Use of AI Tools

Pluto Security Raises $42 Million to Secure Companies’ Use of AI Tools

Startup Pluto Security has raised $42 million to develop tools that monitor the risks of AI tools, agents, extensions, and MCP servers within companies. The company says its annual recurring revenue has exceeded $2 million and that it has more than 30 enterprise customers.

CN
How to Configure GitHub Copilot to Understand Your Software Project’s Rules

How to Configure GitHub Copilot to Understand Your Software Project’s Rules

GitHub explains four ways to give Copilot more precise context within a code repository: custom instructions, agent skills, custom agents, and MCP servers. The goal is to reduce guesswork and enable the tool to follow project standards and test its interfaces using external tools such as Playwright.

CN
FakeGit Campaign Returns with 17,610 Malicious GitHub Repositories to Spread Information-Stealing Malware

FakeGit Campaign Returns with 17,610 Malicious GitHub Repositories to Spread Information-Stealing Malware

The FakeGit campaign has resumed activity through 17,610 fake GitHub repositories, using convincing README files and download buttons to distribute SmartLoader and then other malware. Researchers say that relying on the removal of specific repositories or links is insufficient to contain the campaign because of backups and multiple hosting locations.

CN
Google Moves Gemini into the Agentic AI Phase, Initially Focusing on Businesses

Google Moves Gemini into the Agentic AI Phase, Initially Focusing on Businesses

Google announced a unified agent inside Gemini that can plan tasks and execute them across enterprise applications and systems, with the ability to delegate work to sub-agents and use models from Google and other providers. The rollout will begin with enterprise customers before the technology is offered to consumers, with an independent enterprise identity and an audit log of actions.

CN
ShadowX Develops a Turkish Platform for Enterprise AI Security

ShadowX Develops a Turkish Platform for Enterprise AI Security

ShadowX combines data protection, monitoring the use of AI tools within organizations, and securing models and agents in a single platform that can be deployed on-premises or in the cloud. The company targets the banking, government, and technology sectors, where sharing sensitive data with AI tools poses a growing regulatory and security challenge.

CN
AI Week: The Agent Race Accelerates, While Safety and Funding Questions Expand

AI Week: The Agent Race Accelerates, While Safety and Funding Questions Expand

Webrazzi reviewed a selection of the most notable artificial intelligence developments of the week, from incidents involving agent behavior and the voluntary safety agreement in the United States to new models and products that perform real tasks on behalf of users. Growing investments and infrastructure commitments reveal that the shift from chatbots to agents is simultaneously raising the ceiling for capabilities, costs, and risks.

CN
Why Do Organizations Need Decision-Grade Knowledge in the Age of AI?

Why Do Organizations Need Decision-Grade Knowledge in the Age of AI?

Stack Overflow argues that information retrieval is no longer sufficient for building reliable answers in organizations, especially when sources conflict or become outdated. It presents the concept of “decision-grade knowledge,” which combines evidence, context, permissions, and human review—the foundation the company says Stack Internal is building.

CN
How Forter Enabled Around 200 Employees to Build AI Agents in Two Weeks

How Forter Enabled Around 200 Employees to Build AI Agents in Two Weeks

Ben Maraney of Forter presents a practical experiment in scaling AI-agent development across research and development teams, using a centralized MCP server, multiple platforms for experimentation and execution, and reduced legal and security barriers. The experiment shows that tool accessibility and transparency of tool calls can sometimes matter more than adopting a complex retrieval-augmented generation architecture or advanced evaluation systems from the outset.

CN
What Do the RAG, MCP, and Skills Discussions Tell Us About AI Software Development?

What Do the RAG, MCP, and Skills Discussions Tell Us About AI Software Development?

A GitHub Blog post breaks down five common assumptions about AI-powered software development, emphasizing that reviewing generated code, retrieval-augmented generation, MCP, and Skills are not competing alternatives but tools serving different roles. The conclusion is that human judgment and code maintainability remain decisive factors.

CN
Launch of a Unified Platform for Exploring United Nations System Data with AI

Launch of a Unified Platform for Exploring United Nations System Data with AI

The United Nations system has launched the UN System Data Commons platform, built on Google’s Data Commons, to integrate global statistics into a knowledge graph that can be used with artificial intelligence tools. The platform enables natural-language search and plans to add data from more UN entities, with the goal of reaching 80% of its statistical collections by 2027.

CN
Establishment of the Agentic AI Foundation to Support Open Standards for AI Agents

Establishment of the Agentic AI Foundation to Support Open Standards for AI Agents

The Linux Foundation announced the Agentic AI Foundation, an open-source project focused on developing the infrastructure and standards needed to operate AI agents, with more than 250 companies and organizations joining, along with projects such as MCP, goose, and AGENTS.md.

CN
Google Enables AI Agents to Control Google Home Devices

Google Enables AI Agents to Control Google Home Devices

Google has begun early access to the Model Context Protocol server for the Google Home ecosystem, allowing agents such as Claude and ChatGPT to control connected devices and read camera summaries and home activity history through natural-language commands. The launch is currently limited to paid Advanced plan subscribers in the United States, and the company has not specified when it will expand to other markets or plans.

CN
Kotlin Toolchain 0.12 Expands Multiplatform Library Publishing and Supports Wasm Applications

Kotlin Toolchain 0.12 Expands Multiplatform Library Publishing and Supports Wasm Applications

JetBrains has released Kotlin Toolchain 0.12.0 with experimental support for publishing Kotlin Multiplatform libraries, building Wasm applications for the web, and running Compose Hot Reload from the command line. The release also raises the minimum requirements to JDK 17 and Kotlin 2.2.20, while Toolchain remains in the Alpha stage.

CN
CLion Adds an AI Skill to Analyze Hard Fault Errors in Cortex-M Systems

CLion Adds an AI Skill to Analyze Hard Fault Errors in Cortex-M Systems

JetBrains has added to CLion 2026.2.2 an AI skill that uses MCP tools to analyze fault logs, memory, and instructions when a debugging session stops on a HardFault, with support for multiple debugging tools. The feature is enabled by default, but its support differs between AI modes and agents.

CN
Ibtikar Fund Invests in Palestinian Churn Solution to Develop Subscription Customer Retention Tools

Ibtikar Fund Invests in Palestinian Churn Solution to Develop Subscription Customer Retention Tools

Ibtikar Fund has made an initial investment in Churn Solution, a Palestinian startup that uses artificial intelligence to help subscription companies reduce customer churn and recover lost revenue. The value of the investment, which will fund product development, business expansion, and hiring, was not disclosed.