AI-assisted software delivery is moving beyond individual code suggestions. The next frontier is coordinated execution: understanding the work, organizing the right expertise, respecting dependencies, and validating outcomes before change reaches production.
With the release of Cortex Planner, Cortex adds its ninth specialized model.
Cortex Planner 1.8 brings a new layer of intelligence to the Cortex experience. It transforms a development request into a clear, governed execution plan—helping teams move from “what needs to happen” to “how the work should be coordinated” with greater confidence, visibility, and control.
From routing work to planning outcomes
Cortex Router established a smarter way to direct work across Cortex’s specialized AI capabilities. It evaluates a request and helps identify the right path and the right expertise for the task.
Cortex Planner extends that vision.
Where routing determines where work should go, planning defines what must happen next. Cortex Planner interprets the objective, available context, and permitted actions to create an organized plan for analysis, implementation, testing, security review, integration, and validation.
This is an important step in the evolution of AI-assisted development. Complex engineering work is rarely a single action. A request to add a feature, resolve a defect, investigate a security concern, or improve an existing service may involve several distinct activities—some independent, some sequential, and some requiring careful review before they can move forward.
Cortex Planner helps make that work visible, intentional, and easier to govern.
A more deliberate Cortex experience
Cortex Planner advances Cortex from intelligent task direction toward intelligent orchestration.
Rather than treating a broad request as one opaque operation, Cortex can represent the work as a structured plan with defined stages, specialist responsibilities, and expected outcomes. The planner can distinguish when more discovery is needed from when there is sufficient context to proceed. It can identify tasks that can move forward independently, bring their results together at the appropriate point, and include validation as part of the work from the beginning.
For example, a multi-area product change may require separate implementation, testing, documentation, integration, and validation activities. Cortex Planner can organize those streams of work into a coordinated progression, helping ensure that important branches do not get overlooked and that the final result is assessed against the original objective.
For security-sensitive work, the planner can help structure remediation alongside regression testing, review, integration, and validation. The goal is not simply to produce a proposed fix, but to support a more complete and accountable path to a reliable outcome.
This creates a more deliberate development experience:
- Clearer decomposition of complex engineering work
- Better coordination across specialized AI capabilities
- More intentional sequencing of changes, reviews, tests, and validation
- Greater visibility into what must occur before work is considered complete
- A stronger foundation for governed AI-assisted delivery
Planning with boundaries built in
Powerful AI workflows need more than speed. They need boundaries.
Cortex Planner is designed to operate within trusted context and defined permissions. It does not treat uncertainty as an invitation to take unchecked action. When additional information is needed, it can organize discovery before proposing execution. When changes are being planned, it keeps the work aligned with the available project context. When activities could overlap or conflict, it can organize a safer sequence.
This is especially important as organizations look to apply AI across larger and more consequential software initiatives. Teams need to know not only that AI can generate output, but that it can participate in a process designed around accountability, scope, and validation.
Cortex Planner helps turn those requirements into a natural part of the development workflow.
Designed for specialist collaboration
Cortex is built on the belief that specialized intelligence produces better outcomes than a one-size-fits-all model.
Different engineering activities call for different strengths. Analysis requires investigation and context. Implementation requires focused execution. Testing requires a separate perspective on behavior and quality. Security review requires attention to risk. Integration and validation ensure that individual changes become a coherent, dependable outcome.
Cortex Planner acts as a coordination layer between intent and execution. It helps bring the appropriate specialist capabilities into the appropriate stage of the work, while preserving a clear view of dependencies, acceptance criteria, and completion.
That matters because modern engineering tasks rarely end with a code change. A durable outcome may require understanding an existing implementation, updating multiple areas of a project, adding or improving tests, reviewing risk, integrating the work, and proving that it meets the original objective.
Cortex Planner makes that full lifecycle more visible—and helps teams keep it connected from the first request through final validation.
Making validation part of the plan
One of the most important characteristics of effective software delivery is that validation is not a final checkbox. It is part of how work is designed.
Cortex Planner reflects that principle by incorporating acceptance criteria and validation into the execution plan itself. The intended result is not merely “a change was made.” It is that the requested behavior has been addressed, important constraints have been preserved, and the outcome has been evaluated through the right checks.
For teams, this creates a clearer bridge between a request and a verifiable result. It also helps support more consistent collaboration between developers, security teams, engineering leaders, and the AI capabilities supporting their work.
By making validation visible earlier, teams can spend less time reconstructing intent at the end of a task and more time building confidence throughout it.
Human control where it matters
Automation should increase confidence, not reduce oversight.
Cortex Planner recognizes that some actions require explicit human involvement. When work involves consequential external actions, the plan can preserve a confirmation point rather than assuming authority to proceed. This enables organizations to expand AI-assisted workflows while keeping people in control of decisions that matter.
The result is a practical model for enterprise AI: intelligent enough to organize complex work, structured enough to respect boundaries, and transparent enough to support responsible adoption.
The next chapter for the Cortex AI Ensemble
Cortex Planner is the ninth model in the Cortex AI Ensemble and a meaningful expansion of what Cortex can help engineering organizations coordinate.
It builds on the progress of Cortex Router by adding a planning layer that connects intent, specialist collaboration, governed execution, and validation. It helps shift the conversation from isolated AI interactions to coordinated AI-assisted delivery—where each contribution is part of a larger, understandable path toward an outcome.
The future of software engineering will not be defined only by how quickly AI can generate code. It will be defined by how effectively AI can help teams organize complex work, preserve quality, manage risk, and deliver outcomes they can trust.
Cortex Planner brings Cortex closer to that future: a coordinated AI experience where every request has a clearer path forward, every specialist contribution has a purpose, and every outcome is designed to be understood, validated, and trusted.
