Google has enhanced its Gemini CLI extension, Conductor, by adding support for automated reviews. The company says this update allows Conductor “to go beyond just planning and execution into validation”, enabling it to check AI-generated code for quality and adherence to guidelines, strengthening confidence, safety, and control in AI-assisted development workflows.
Our new Automated Review feature introduces a rigorous “verify” step to the development lifecycle. Once your coding agent completes its tasks, Conductor can now generate a comprehensive post-implementation report.
Google says the new feature automatically evaluates AI-generated code across multiple dimensions, including code quality, plan compliance, enforcement of style and programming guidelines, integrated test validation, and basic security review.
For code quality, Conductor can now perform deep static and logic analysis beyond syntax, flagging complex issues such as race conditions, null pointer risks, and logic bugs. It also verifies that the actual code implementation matches the original plan.md and spec.md, ensuring all planned tasks are properly addressed.
Conductor also ensures that code adheres to your project’s style guides and any custom rules defined during the planning phase. It automatically runs unit and integration tests, including test results and coverage data in the review report. Finally, it scans for critical security issues before code is merged, such as hardcoded API keys, potential PII leaks, or unsafe input handling.
The report generated by Conductor categorize findings by high, medium, and low severity, including exact file paths and the ability to start a track to fix issues.
This level of detail ensures that “agentic” development doesn’t mean “unsupervised” development. Instead, it creates a workflow where the AI provides the labor and the developer provides the high-level architectural oversight, backed by automated verification.
Google says that automated reviews in Conductor “close the loop” in AI-assisted engineering by adding validation and verification after code is written. The feature helps address growing concerns about AI-generated code quality, security, and compliance, aiming to make AI-driven development safer, predictable, and architecturally sound.
Launched last December, Conductor is an extension for the Gemini CLI that supports context-driven development by storing project context, including plans, specs, and style guides, in version-controlled Markdown files rather than ephemeral chat logs.
In Conductor, a track represents a discrete unit of work with a written specification and a task-oriented plan divided into phases and subtasks. Implementation begins only after plan review, with progress tracked directly in the plan file.
Conductor supports team-wide configuration, letting projects define shared standards—like testing strategies, coding conventions, and workflow preferences—once and apply them consistently across all AI-assisted contributions, ensuring consistency for both individual developers and teams.
