Qodo launches governance tools for AI code reviews

Qodo has launched three governance tools for AI-assisted software development, targeting code review and standards management across enterprise engineering teams.

The new products are Cross-Repo Code Review, Custom Rules Miner and Skill Review Standards. They are designed to address governance problems that arise as AI agents generate and submit more code across the software development process.

The company's argument rests on a shift in how software teams work. Instead of developers writing and reviewing code at a human pace, AI agents are increasingly producing changes across multiple repositories, leaving existing quality controls struggling to keep up.

Qodo cited Google DORA 2025 research showing that pull requests from teams with high AI adoption are 154% larger, take 91% longer to review and ship 9% more bugs. The figures suggest a growing gap between the volume of code being produced and engineering teams' ability to review it thoroughly.

Three problems are becoming more common in large organisations: cross-repository dependencies can fail when ownership is fragmented between teams; engineering knowledge is often scattered across documents and review history; and standards embedded in agent workflows do not always carry through into the review process.

Review across repos

Cross-Repo Code Review is aimed at software changes that extend beyond a single code repository. In many large engineering environments, updates to shared libraries, application programming interfaces, data schemas or infrastructure files can trigger downstream problems in connected services without those issues appearing during a standard pull request review.

The beta feature extends Qodo's Git plugin so that when a pull request changes a shared dependency, the system reads registered consumer repositories and flags potential impact before the change is merged. Findings can include function signature violations, broken API contracts, schema changes and infrastructure drift.

This gives engineering teams an assessment of likely cross-system effects within the normal review workflow. The aim is to catch issues that might otherwise surface only after code has been merged and deployed.

Mining standards

Custom Rules Miner addresses a different problem: many organisations lack a single, machine-readable source of coding standards. In practice, those rules often sit in internal wikis, pull request comments or the judgment of experienced engineers rather than in a format software tools can enforce consistently.

The feature analyses existing codebase behaviour and pull request history to identify recurring coding patterns. It then presents those patterns as structured rules that can be applied within the platform.

This approach means teams do not need to write standards from scratch before applying them. Instead, the system attempts to infer standards from how engineers have already worked and reviewed code over time.

Skills governance

The third launch, Skill Review Standards, focuses on the use of agent skills, which organisations use to encode development workflows, review instructions and internal best practice. As those skills spread across repositories, managing them can become difficult, especially when standards vary between teams or are not centrally tracked.

The feature offers centralised management for skills that contain code review instructions and coding standards. It discovers those skills across repositories, presents them in a dedicated portal and gives teams controls and analytics to monitor their use and effect.

This creates a layer of oversight for standards that might otherwise remain dispersed across files and teams. It also links review standards more directly to the review process itself, an area that has often remained disconnected.

Qodo counts Walmart, NVIDIA, Red Hat and Monday.com among its customers. Founded in 2022, the company has raised USD $120 million from venture investors and angel backers, including executives from OpenAI, Meta, Shopify and Snyk.

The launch reflects a wider debate inside large software organisations about whether governance, rather than code generation itself, is becoming the main operational challenge in AI-led development. As AI tools take on more of the drafting and submission of software changes, companies are under pressure to show that quality, consistency and traceability can still be maintained.

Itamar Friedman, Chief Executive Officer and Co-Founder of Qodo, said the issue has moved beyond conventional development tooling.

"The volume of AI-generated code has outpaced every quality process enterprises had in place. Engineering organizations now need three things they have never had to govern at this scale: standards that exist somewhere a system can read and enforce, agents that apply those standards consistently, and visibility into the health of a codebase that no single engineer can hold in their head anymore. That is not a tooling problem. That is infrastructure," said Itamar Friedman, Chief Executive Officer and Co-Founder of Qodo.