The Linear cloudy issue tracker and project manager has introduced an AI agent and plans to add AI coding assistance, with CEO and co-founder Karri Saarinen declaring that "issue tracking is dead."

Now in beta, the Linear Agent works in the online, mobile, or desktop app, and as a plug-in for products including Slack, Teams, and Zendesk. It has a chat user interface and examples of usage are things like "make issues based on the discussion here and assign them to me," according to the introductory post.

The agent also supports skills and automations, which require a business ($16 per user/month) or enterprise plan. Skills are workflows saved for future use to automate common tasks. Automations are workflows triggered automatically when issues are created.

Future features will include a coding agent to write code and fix bugs, as well as the ability to answer questions about a codebase and to present code diffs (changes made to a block of code).

Pricing is unchanged during the beta, but the post suggests that automations and coding features will move to usage-based pricing "beyond a certain threshold."

Saarinen said that "issue tracking is dead," arguing that agents "make software development a lot simpler" as they do more of the procedural work. He said that coding agents are already installed in 75 percent of Linear enterprise workspaces, and that the volume of work done by agents has increased five times in the last three months.

A diagram in his post shows Linear as a tool for capturing context while agents do most of the engineering work. Despite his post, issues will remain as the core items in Linear.

In the company's Slack channel, customers were positive about the announcement, but asked for model context protocol (MCP) support. An employee responded that "we're working on MCP support."

• JetBrains shifts to agentic dev with Central, retires pair programming
• Mozilla introduces cq, describing it as 'Stack Overflow for agents'
• Avalonia bolts Linux and WebAssembly onto .NET MAUI
• Fixing Claude with Claude: Anthropic reports on AI site reliability engineering

One thing not mentioned in any of the posts or agent documentation is security, other than that "Linear Agent operates within your existing permissions." Generative AI systems can be vulnerable to malicious prompt injection and other errors. This will become a bigger concern as the capabilities of the agent increase, for example, when it is able to write code and interact with MCP servers.

Users will get the new agent by default, but it can be disabled in the settings.

Views on AI's value in software development vary, but it is shaking every aspect of the profession regardless. The problem for services such as Linear is that if coding continues to travel towards more agentic workflows, customers may shift away toward tools designed specifically for agent management.

Basecamp from 37signals, another project management tool, is also planning to reposition itself as "agent first, agent native" with access from any AI agent via a command line interface and the promise of becoming "an agent-driven assistant."

These products all aim to place themselves at the center of an agent-driven future, but they cannot all be in that position. ®

More about

• AI
• Developer
• Software

More like these