LinearB launches Agentic Studio to provide engineering context to coding agents – Unite.AI

LinearB launched Agentic Studio to transform software delivery data into context that technical leaders and coding agents can use to make decisions. Announced on November 2, the platform combines natural language analysis with an engineering context layer that spans the tools teams use to plan, write, review and release software.

The launch reflects a growing problem in AI-assisted development: Producing more code doesn’t automatically help an organization deliver useful software faster. Reviews can become congested, changes can become more difficult to evaluate, and agents can start a business without knowing which services are fragile or why a particular solution is important.

Agentic Studio is available to all LinearB customers, according to the announcement. Agent skills, including the initial ability to match an agent’s effort to a task, are in early access.

Transform delivery data in an engineering context

LinearB collects information during software deployment and normalizes it into a common model. The company says it reconciles engineer identities between Git and project management systems, links changes from ticket to release, and adds attribution and AI benchmarking to help leaders interpret results.

The important step is to link these records. A pull request can seem simple on its own while impacting a service with recent incidents, frequent rework, or complicated properties. Linking the change to that history provides the reviewer or coding agent with information that a snapshot of the repository alone may not provide.

In the new studio, leaders can ask questions in plain language, get suggestions on next steps, and use AI assistance to create metrics, dashboards, and other artifacts. LinearB says organizations can define the context available to users with role-based access built into the platform.

This direction is in line with the company’s guidance on context engineering, which emphasizes selecting relevant information rather than indiscriminately filling an agent’s context window. Its framework considers the code, the intent behind a change, the connections between systems, and the limits of what an agent can access.

An early access skill to match effort and risk

The first agent skill is designed to decide how much effort a coding task requires. LinearB says it considers the relevance, complexity and risk of the business, using signals such as rework rates, recent service incidents and ownership history.

A low-risk change may receive lighter treatment with fewer tokens, while a more consequential change may receive more targeted analysis and testing. Engineering leaders can see how often tasks are routed to low, medium, or high effort. The company says the skill works with any coding agent, although the announcement does not provide comparative results between specific agents or quantify savings made on tokens.

The engineering idea is useful even before a spending request is established. A uniform reasoning budget treats a minor maintenance change and a sensitive production fix equally. The specific context of the task could help focus more attention where mistakes are costly, while reducing unnecessary work on simpler changes.

MCP connects data to existing AI workflows

LinearB also exposes design information via the Model Context Protocol, or MCP. The MCP product page describes natural language access to delivery metrics, custom reports, and bottleneck analysis, along with recommendations such as review automation and routing.

This gives teams the ability to analyze technical data from an AI client instead of manually moving between dashboards. MCP provides connection to tools and information; the context layer provides the delivery-specific relationships that make a question about a repository or team meaningful.

There is a distinction to resolve with the launch of the new platform. While the launch describes role-based access and enterprise guardrails, LinearB’s public MCP FAQ still lists permission-scoped MCP access as a roadmap item. The new study’s access control statement should therefore not be read as confirmation that every existing MCP client already inherits every LinearB permission.

Because more AI production can expose downstream bottlenecks

LinearB’s announcement cites AI 2026 benchmarks drawn from 2.7 million pull requests and 83,000 developers across 253 organizations. According to the company, developers using AI on at least 75% of programming days merged 96% more pull requests per week compared to the previous year.

These figures describe the observed activity, rather than proving that AI alone caused the change. Additionally, a larger number of merged pull requests does not demonstrate that customers received double the value. The release separately cites findings that AI-assisted changes were larger and pull requests created by agents waited substantially longer for review, illustrating how faster code generation can shift work to the rest of the delivery process.

The example of a customer makes this distinction concrete. LinearB says Kraken used its AI-powered MCP server to investigate review delays and found that time zones and reviewer availability matter more than big changes. Kraken reportedly reduced review times at the 90th percentile by more than half in four months. The company’s published discussion with Kraken provides additional context on using repository data and deployment metrics to uncover bottlenecks.

From measuring AI adoption to improving delivery

LinearB plans additional agent skills for downstream work such as review, merger, and governance. These are future directions, distinct from the current studio availability and early access status of the effort ability.

The broader thesis is that coding agents need an understanding of the organization around the code: what matters, who owns it, where previous changes have failed, and what constraints apply. Agentic Studio introduces this proposition to the developer productivity platform, the value of which is ultimately measured by the quality and flow of changes made rather than the quantity of code generated.