For years, software delivery moved at human speed, with changes written, reviewed, and tested by people. AI has removed that constraint—teams now produce more code than ever, increasingly with agents generating changes alongside humans. The systems responsible for testing, validating, and governing software haven’t kept pace, leaving teams with less visibility and control over what they are actually releasing.

Codecov provides critical visibility into how much of an application’s codebase has been tested and whether it meets the threshold required to ship safely. By embedding Codecov’s coverage intelligence directly into the Harness delivery pipeline, coverage signals can inform real-time release decisions before code is shipped to production.

Also Read: 42Crunch AI Coding Plugins Unlock Agentic DevSecOps for the Enterprise

“AI is dramatically accelerating code creation, but software delivery ultimately comes down to deciding what’s safe to release and when,” said Jyoti Bansal, co-founder and CEO of Harness. “By bringing Codecov into Harness, we can connect coverage data directly to delivery decisions so teams can move faster without sacrificing trust in what they ship. The future of software delivery will depend on turning signals like coverage into enforceable decisions that guide every release.”

With Codecov joining Harness, coverage data becomes part of the Harness Software Delivery Knowledge Graph, which connects deployment signals across the delivery lifecycle. Codecov adds a critical new layer to that intelligence: real-time coverage data tied directly to the repositories and changes flowing through the delivery pipeline.

By connecting coverage data with deployment history, security reachability analysis, incident telemetry, and change history, the platform can help teams understand release risk before code reaches production—giving them visibility into what changed, what was tested, and where risk may be accumulating as software delivery continues to accelerate.

Commitment to the Codecov Product and Community

Codecov is trusted by thousands of development teams and widely used across the open source ecosystem. Harness plans to continue investing in Codecov while preserving the experience teams rely on today.

The transition reflects a shared view that modern software delivery depends on strong alignment between development signals, operational visibility, and release governance. Harness plans to accelerate innovation in Codecov while retaining its brand and free access for open source projects.

“Sentry’s mission is to help developers see what’s breaking in production so they can fix it fast. This focus is what shapes every decision we make,” said Milin Desai, CEO, Sentry. “Codecov has become an important tool for thousands of engineering teams, and it deserves the investment and roadmap focus to match that. Coverage intelligence works best when it’s embedded in the delivery pipeline, and that’s exactly where Harness operates. We didn’t just find Codecov a home, we found it the right one.”

Ongoing Partnership Between Harness and Sentry

Harness and Sentry are committed to ensuring continuity for Codecov users while building long-term momentum for the product. The companies share a belief that the future of software delivery requires closer alignment between quality, reliability, and visibility.

Together, Harness and Sentry will continue supporting developers and engineering teams through integrations, developer community engagement, and shared efforts to advance the tools and practices needed for the AI-accelerated future of software delivery.

Source: PRNewswire

The post Harness Acquires Codecov from Sentry to Strengthen Software Delivery Governance in the AI Era appeared first on ITDigest.

As AI agents increasingly write and modify code, the pace of software development is accelerating beyond the limits of traditional security processes. APIs – the backbone of modern applications – are being created, deployed and consumed at machine speed, expanding the attack surface and exposing a growing gap between development velocity and security oversight. The 42Crunch API security platform plugin for Claude Code closes that gap.

The integration enables real-time detection and automatic remediation of API vulnerabilities introduced during AI-driven development. Rather than stopping at identification, 42Crunch generates context-aware fixes, applies them automatically, and re-tests the updated code—creating a continuous detect-and-fix loop that operates without human intervention.

Also Read: OpenAI Accelerates Agentic Workflows with WebSockets to Unlock Real-Time AI Performance

The result is a shift from periodic, manual security checks to continuous, automated enforcement embedded directly in development workflows.

Rik Turner, Chief Analyst at Omdia, said: “Agentic AI is fundamentally reshaping how applications are built, with APIs increasingly generated and consumed at machine speed. This creates a widening gap between development velocity and traditional security practices. By integrating real-time detection and automated remediation into AI-driven workflows 42Crunch is executing a critical step toward enabling enterprises to scale AI adoption securely.”

Jacques Declas, CEO of 42Crunch, said: “As AI agents take on a greater role in writing and executing code, API security becomes the control layer and it must operate in real time at the same speed. Security leaders can no longer rely on reactive, manual approaches. Security has to be continuous and automated and that’s exactly what this integration delivers.”

“With 42Crunch and Claude Code, security is embedded across the development lifecycle, from design through runtime, without slowing development. That brings the industry closer to a practical realization of agentic DevSecOps at scale,” added Declas.

Source: BusinessWire

The post 42Crunch AI Coding Plugins Unlock Agentic DevSecOps for the Enterprise appeared first on ITDigest.

The automation devices running the world’s plants, data centers, and infrastructure don’t just produce output. They produce the U.S. economy. The code controlling them has quietly become as critical as the financial systems and enterprise software that get the security headlines, and yet most of that code is still managed by hand, with risk surfacing only when something breaks or a regulator asks for proof.

That risk is no longer hypothetical. In April 2026, a foreign threat group targeted manufacturing devices across the U.S., a reminder that industrial code is now a frontline target. Copia’s platform was built for this moment. Copia Source Control gives teams a verified, auditable record of every change. DeviceLink keeps a clean copy of what’s running on every device. And now, with Copia Actions, OT teams can govern and standardize that code automatically, ensuring that when an incident occurs, recovery to a verified, known-good state takes minutes rather than days.

Also Read: Moderne Introduces Changelog to Bridge Source Control and Change Control for AI-Driven, Multi-Repository Software Engineering

This is the discipline software engineering has practiced for nearly two decades. When a software developer saves a change today, dozens of automated checks run before that change reaches production: tests, security scans, compliance reviews, all triggered automatically and tied to the specific change that prompted them. Industrial engineering has never had an equivalent until Copia Actions. Purpose-built for PLC, HMI, and robotic-controller code, it is configurable to the standards each organization has always intended to enforce, and extensible as those standards grow.

“Industrial code now determines audit posture, cyber resilience, and time-to-recovery for the world’s manufacturers, infrastructure, and data centers, yet most of it is still managed manually,” said Adam Gluck, founder and CEO of Copia Automation. “Copia Actions closes that gap. It gives engineering and security leaders the same automated governance that IT organizations have relied on for nearly two decades.”

The launch of Copia Actions reflects a commitment Copia has held since its founding: bringing the engineering disciplines that transformed software development to the industrial code running the physical world. With Source Control, DeviceLink, and now Actions, Copia continues to close the gap between how OT and IT manage their most critical assets, and to push the industry toward the standards its customers, regulators, and operators increasingly demand.

Source: PRNewswire

The post Copia Automation Launches Copia Actions: The First Configurable Workflow Engine Purpose-Built for Industrial Code appeared first on ITDigest.

The post Sonar Acquires Gitar to Strengthen AI Code Verification and Review Capabilities appeared first on ITDigest.

Engineering leaders running large initiativces lack a single view into what is changing across their software estate: which pull requests are open, what is blocked, and where work is being duplicated across repositories. Source control systems are designed to answer one question: what changed inside this repository? Moderne Changelog answers a different one: what is changing across the organization, and is it coordinating properly? Changelog operates as a unified control plane that overlays existing source code management (SCM) systems, including GitHub organizations, Bitbucket instances, GitLab deployments, and internal SCM systems spread across separate network segments, delivering a portfolio-wide view of change scoped to organizational structure as defined in the Moderne Platform.

Engineering teams can see the full state of change across every repository in their portfolio, filter across any dimension of that work, and act directly, approving, merging, or closing batches of pull requests in bulk. Not all changes carry the same review expectations: manual edits, agent-driven modifications, and automated transformations each require different levels of scrutiny. Changelog preserves that distinction, giving reviewers immediate visibility into which parts of an initiative came from automation and which require human judgment, without requiring them to navigate repository by repository. For changes generated by Moderne’s transformation recipes, results surface in Changelog as a changeset, grouping all affected repositories into a single coordinated effort.

Also Read: GitLab Expands Agentic AI Capabilities to Automate Security, Pipelines, and Delivery Insights

For AI coding agents, the absence of cross-repository visibility carries direct operational costs. Without it, agents duplicate in-progress work, sequence changes incorrectly across dependency boundaries, and burn tokens reconstructing context that already exists. In a repository estate where a shared library must build and publish before dependent services can be safely updated, an agent with no cross-repository view has no way to know whether that prerequisite has been met. Changelog exposes what is ready, what is blocked, and what is waiting across the entire changeset through Moderne’s MCP interface and GraphQL API, so agents maintain operational context rather than starting from scratch each time, inheriting existing SSO and SCM permissions with no new access grants required.

“Every time an agent scans a repository to reconstruct what’s already in flight, it’s burning tokens on a problem that better tooling can solve,” said Jonathan Schneider, CEO and co-founder of Moderne. “Changelog gives agents and the engineers working alongside them a direct feed of what’s changing across the organization. That’s real compute you get back.”

Moderne Changelog is the latest addition to Moderne’s agent tools, which includes Prethink for compiler-accurate codebase context and Trigrep for high-speed, symbol-aware code search. Together, these capabilities give engineering teams and AI agents the infrastructure to understand large codebases, search them precisely, and coordinate change across them all surfaced through Moderne’s MCP interface and grounded in OpenRewrite, the open-source transformation engine Moderne created and maintains.

Source: BusinessWire

The post Moderne Introduces Changelog to Bridge Source Control and Change Control for AI-Driven, Multi-Repository Software Engineering appeared first on ITDigest.

Also Read: UiPath and Deloitte Partner to Advance Agentic AI-Powered Software Testing

Moreover, having GitLab in the Claude Marketplace will allow the company’s customers to leverage their Anthropic spending commitment towards any agentic uses of AI within the software development lifecycle process. Since AI agents will continue to perform tasks like planning, coding, testing, and deploying applications, the availability of the marketplace offering will ensure that security and compliance teams will have visibility into their activities and remain in control of the process.

Read More: GitLab Deepens Integration with Anthropic’s Claude Models to Accelerate Secure Software Development

The post GitLab Expands Claude Integration to Advance Secure, Governed AI-Driven Development appeared first on ITDigest.

Also Read: GitLab Expands AWS Integration to Power Agentic DevSecOps with Amazon Bedrock

The tool will allow businesses to concentrate on innovation rather than routine maintenance. UiPath Autopilot for Testers and Agent Builder are also included in the software, giving companies access to more than 1,500 pre-built bots and AI agents specific to certain domains. “This collaboration redefines how testing is done in the AI era, where AI agents and testers work together to deliver speed, resilience, and innovation at scale,” said Gerd Weishaar, SVP of Product Management for UiPath Test Cloud.

Read More: UiPath Accelerates New Agentic-led Software Testing Capability through Deloitte’s ASCEND Delivery Platform

The post UiPath and Deloitte Partner to Advance Agentic AI-Powered Software Testing appeared first on ITDigest.

Agentic workflows are used in applications like coding assistants and automation tools and they comprise many iterations of AI model and external tool interactions. Previously each interaction physically required making a new HTTP request, which contributed additional latency and made the whole process slower. The new OpenAI’s methodology gets rid of it by using a WebSocket connection that is persistent and allows continuous real-time communication between client and the API.

This change in architecture allows the system to store the context and the use of an earlier response without the need for reprocessing while also decreasing the network overhead. Therefore, according to OpenAI, there are up to 40 percent faster end-to-end performance in agentic workflows as a result of combined improvements in responsiveness and time-to-first-token.

This is a crucial improvement as AI models are getting faster and more capable. Model inference was the main performance bottleneck in the past. Yet, with the current generation of models that can produce tokens at break-neck speed, the whole supporting infrastructure, most notably the API communication, has become the bottleneck. By enhancing this part, OpenAI is helping the developers in making full use of the speed of the latest AI systems.

Implications for the IT Industry

OpenAI’s WebSocket incorporation is indicative of a larger movement in the IT sector which is leaning toward real-time, always-on artificial intelligence infrastructure. As agentic AI systems are becoming more advanced, traditional request-response based architectures are no longer sufficient to manage continuous, multi-step workflows.

In the light of this advancement, IT personnel need to begin to plan their system designs around low-latency, stateful communication models. Persistent connections, in-memory caching, and asynchronous processing are some of the features which together form the key building blocks of contemporary AI architectures.

This feature change also draws attention to the rising significance of infrastructure optimization with regard to AI implementation. We have seen that while innovations in models still attract most of the focus, the performance improvements are now equally hinged on the effectiveness of the supporting ecosystemAPI, networking, and orchestration layers.

Besides, enabling quicker agentic workflows lead to a shorter timeline for the introduction of autonomous AI systems in various sectors. Whether it is a case of software creation or customer care automation, businesses can introduce agents that can work in a more integrated manner and give responses almost in real time.

Also Read: Chainguard and Cursor Partner to Secure Agentic Coding with Trusted Open Source

Business Impact and Strategic Value

Increased speeds in the agentic workflow process can result in better efficiency and end-user experience for corporations. For example, software applications based on AI technology like coding aids, analytical programs, and customer service modules can generate faster results.

Additionally, increased speed also allows for the scaling of AI-based processes more efficiently. Since AI processes become faster and can achieve higher throughput levels, companies can perform more functions without requiring additional expenses on infrastructure development.

In addition, reduced latency also results in the increased reliability of AI algorithms in critical applications. Latency-free processing plays an important role in certain applications such as financial modeling, real-time monitoring, and autonomous decision making.

Finally, the transition to persistent and efficient communication frameworks also leads to lower operational expenses because fewer calculations and network interactions are required.

Advancing the Future of Agentic AI

OpenAI’s WebSocket-based enhancement marks a pivotal step in the evolution of agentic AI, emphasizing that speed and efficiency are as critical as intelligence. As AI systems become more autonomous and deeply integrated into business operations, the ability to execute workflows quickly and seamlessly will be a key differentiator.

For the IT industry and businesses alike, this development signals a future where AI is not only more capable but also more responsive and scalable. Organizations that adopt optimized, real-time AI infrastructure will be better positioned to harness the full potential of agentic workflows and drive innovation in an increasingly AI-driven landscape.

The post OpenAI Accelerates Agentic Workflows with WebSockets to Unlock Real-Time AI Performance appeared first on ITDigest.

Also Read: GitLab and Google Cloud Partner to Deliver Agentic DevSecOps with Vertex AI Integration

The integration also builds on GitLab’s Bring Your Own Model (BYOM) capability, allowing self-managed customers to connect directly to Amazon Bedrock while keeping source code and inference traffic within their network boundaries. GitLab credits bought from AWS Marketplace are also added to their AWS spending, thus simplifying scaling up the AI efforts for them. In conclusion, this collaboration will help organizations leverage DevSecOps through agency, efficiency, and compliance.

Read More: GitLab Collaborates with AWS to Bring Agentic DevSecOps to Enterprise Teams Using Their Existing Amazon Bedrock Accounts and Spend

The post GitLab Expands AWS Integration to Power Agentic DevSecOps with Amazon Bedrock appeared first on ITDigest.

These new features add AI agents capable of detecting issues in the code on their own, suggesting changes, and sometimes even fixing problems automatically. This is a significant change in security practices – typically, problems would be identified but it would be up to human intervention to address them – now the focus is on autonomous and actionable approach. In fact, GitLab achieves this by integrating the capabilities right into the development pipeline, so that risk is mitigated without slowing down innovation.

Besides improving security measures, GitLab also released features which can automate pipeline setup. These enable development groups to instantly set up CI/CD pipelines with the help of AI recommending the best practices and project needs. As a result, the complexity of getting new projects on board is minimized and the consistency of the development workflows is maintained.

Furthermore, the platform has upgraded the delivery analytics that gives teams a comprehensive understanding of the software performance, deployment frequency, and bottlenecks. By using these analytics, business organizations can rely on data to help them make decisions and at the same time, optimize Prouction continuously. On the whole, these changes illustrate GitLab’s commitment to being a single DevSecOps platform with the support of smart automation.

Also Read: NVIDIA Launches Ising Open AI Models to Accelerate the Path to Practical Quantum Computing

Implications for the IT Industry

GitLab’s expansion of agentic AI capabilities reflects a broader transformation in the IT industry toward autonomous software development and operations. Software systems are getting more complex and releases cycles faster, making the traditional manual processes insufficient to maintain the speed and security.

Introducing AI-driven security remediation is very important. The volume and sophistication of cybersecurity threats are both increasing. Most of the time organizations are incapable of vulnerability management. Automating detection and remediation through platforms such as GitLab is a great way to transform security from being a reactive function to a proactive and integrated part of development.

Automated pipeline setup is a step further to emphasize developer productivity and standardization. IT teams could reduce configuration overhead, minimize errors and accelerate time-to-market by using AI-driven automation.

Besides that, with delivery analytics, the industry is also focusing on data-driven DevOps practices. Nowadays, organizations are relying on constant insights to optimize their performance, improve collaboration and deliver better results.

Business Impact and Strategic Value

A number of these new features of GitLab present great advantages to enterprises such as a powerful blend of swiftness, security, and productivity. Automation can easily help in ways of organizing code related pipelines, detecting and fixing vulnerabilities. Through these measures, organizations can operate at lower cost and developers will have more time on their hands for activities that contribute to their personal development as well as the company.

Security is next on the list of benefits. For example, automated remediation minimizes the chances of a fatal breach, non-fulfillment of compliance, among others. In such cases, IT keeps business data safe and customer’s trust is gained and retained.

At the same time, these reformed analytics execution modes serve the purpose of facilitating good decisions making. Companies derive development pattern loopholes and sort out the allocation of their resources in efficient manner so that final outputs meet their time and financial plans.

What is more, software quick and steady delivery can in turn be a source of supply advantage especially for ones belonging to industries where it is innovation and speed that count the most. These are the kind of organizations whose positioning in the market is the most enabling in spite of the extreme and mobile nature of customer demands to which they have to always respond. Moreover, they are able to launch their new and innovative apps with the highest quality.

Advancing the Future of DevSecOps

GitLab’s latest updates highlight a defining trend in enterprise technology: the evolution of DevSecOps into intelligent, self-optimizing systems powered by agentic AI. As AI continues to mature, development platforms are becoming more proactive, capable of not only identifying issues but also resolving them autonomously.

For the IT industry and businesses alike, this marks a new era where software development is faster, more secure, and increasingly automated. Organizations that embrace these advancements will be better equipped to navigate the complexities of modern IT environments and drive sustained innovation in an AI-driven world.

The post GitLab Expands Agentic AI Capabilities to Automate Security, Pipelines, and Delivery Insights appeared first on ITDigest.