Harness Launches Code Repository With AI Code Review for Agent-Ready Development

Built for the volume of AI-generated code, with permissions for AI agents, reviews that keep pace, an enterprise free plan, and one-click migration from existing Git providers

SAN FRANCISCO, Aug. 27, 2026 /PRNewswire/ -- Harness, the platform for the autonomous SDLC, today launched Agent-Ready Harness Code Repository and AI Code Review, built for teams increasing their adoption of AI coding agents. The capabilities are designed to work as one experience: code lands in a repository built for the volume now generated by agents, and is reviewed by a system that understands what a change puts at risk.

Coding stopped being the hard part

Generating code is no longer the constraint. With agents in the loop, teams produce far more code, far faster, than they could alone, and the work has shifted to what happens next: storing, reviewing, approving, and shipping all of it without anything breaking.

The tools underneath were not built for that. Legacy source code management tools in wide use today, GitHub included, assume a human writes code and opens a pull request before a few colleagues look at it over the next few hours or days. When agents are doing the writing, the cracks show up in ordinary places: search and file history get slower as indexing falls behind, pull requests pile up faster than anyone can read them, and a permission system designed around a list of developers has no good answer for an agent that might merge code on its own. Harness has now rebuilt that layer.

"Software delivery is going through its biggest shift since the move to the cloud, and the systems we all built our workflows around were designed for a different scale and a different kind of user," said Jyoti Bansal, CEO and co-founder of Harness. "You do not solve that by adding AI features to a repository designed fifteen years ago. The entire SDLC has to become autonomous, which means the repository, the review, the pipeline, and the governance must all work as one system."

Inside Agent-Ready Harness Code Repository

Harness Code Repository is what source control looks like when agents are part of the team.

• Keeps up as commit and pull request volume increases: Harness Code Repository is scale-tested to handle thousands of pull requests and commits opened at once, which is roughly what a team running agents looks like on an ordinary day. Search, history, and diffs keep working at that volume, across repositories of varying sizes and with branches in the tens of thousands.
• Agents get their own permissions: Agents inherit permissions from humans that trigger them. Developers can further define what an agent may touch, merge, or deploy, just as they would scope a new engineer, down to a specific repository, branch, or environment. RBAC and Open Policy Agent (OPA) policies put that boundary in writing rather than leaving it to convention, so it holds before the merge instead of turning up in an audit afterward.
• Tailored to be used by humans and agents via MCP and CLI: Harness Code Repository and AI Code Review are both reachable from the command line through the Harness MCP and CLI, which covers the full pull request lifecycle, allowing users to find a review by the author's email instead of an internal ID, see every open pull request across every repository in one place, and create, reply to, or resolve comment threads without opening a browser. AI agents can leverage Harness CLI commands to execute Harness Code and Review operations more deterministically and in headless mode.
• Free to start, simple to migrate: Any team can start using Harness Code Repository for free, with 50 GB storage. Migration takes a few clicks, regardless of whether organizations are migrating one repository or an entire GitHub organization, GitLab group, Bitbucket workspace, or Azure DevOps project, and the Harness CLI brings pull requests, labels, webhooks, and branch rules along with it.

AI Code Review

A repository that can withstand a flood of code does not help if a human still has to review all of it. AI Code Review reads a pull request the way a tech lead would.

• Checks that gate the merge. Teams decide which AI Checks are mandatory, setting them once for an account or tuning them by project, and a change that fails a required check cannot be merged.
• Diff grouping by risk. The diff is grouped by risk rather than by file, so the few high-risk changes that alter how the software behaves are not buried under mass renames and dependency bumps.
• One-click remediation. Feedback describes what a change puts at stake rather than noting that a line moved, and suggested reviewers and labels arrive before anyone opens the pull request. If the feedback is valid, changes can be merged with a single click.

Agents can write code, but someone still has to decide what ships. AI Code Review is built so that the decision is an informed one instead of a rushed one.

One platform, from commit to production

Committing, reviewing, building, testing, securing, and deploying code already follow a single sequence inside Harness, which is why the repository and the review layer could ship together. Both are now part of the outer loop that the Harness Software Delivery Agent runs end-to-end, from commit to production, under one policy engine.

That sequence is also what makes the review useful. Harness already knows how a team releases software, which policies they enforce, and what has failed in their production before. All of that is mapped in the Harness SDLC Knowledge Graph, providing enterprise context for every review to reference.

By dogfooding these capabilities over the past several months, Harness engineering teams have saved more than 10,000 hours of manual review time per month.

Availability

Agent-Ready Harness Code Repository and AI Code Review are available today – book a demo to learn more.

About Harness

Harness is the AI Software Delivery Platform™ company, enabling engineering teams to build, test, and deliver software faster and more securely. Powered by Harness AI and the Software Delivery Knowledge Graph, the platform brings intelligent automation to every stage of the software delivery lifecycle after code — removing toil and freeing developers from manual, repetitive work. Companies like United Airlines, Morningstar, and Choice Hotels use Harness to accelerate releases by up to 75%, cut cloud costs by 60%, and achieve 10x efficiency across DevOps. Based in San Francisco, Harness is backed by Goldman Sachs, Menlo Ventures, IVP, Unusual Ventures, and Citi Ventures.

SOURCE Harness