Erfan Al-Hossami
Yiwen Xu
June 11, 2026
4 min read
June 11, 2026
4 min read
Share
Cut code review time & bugs by 50%
Most installed AI app on GitHub and GitLab
Free 14-day trial
A pull request can look safe in one repository and still break another service that depends on it. That is why we shipped Multi-Repo Analysis: to help CodeRabbit review changes with context from related repositories. The challenge was set up. Teams still had to tell CodeRabbit which repositories were connected, and that list could go stale as services, packages, and ownership changed.
Automatic Repository Linking removes that manual step. CodeRabbit can now detect related repositories across your organization and use them as review context, so cross-repo impact is easier to catch before you merge.
Automatic Repository Linking discovers and links related repositories across your organization. It first analyzes signals that define a dependency such as import graphs, dependency manifests, and shared code patterns. It then links the repositories that depend on each other, so CodeRabbit knows your architecture without anyone maintaining a list.
What that gives your team:
Cross-repo context with zero setup. Related repositories get linked automatically. You do not need to maintain a static list of related repositories as services and packages evolve.
Links that reflect how your code fits together. Because the connections come from the code itself, they map to real technical dependencies and stay current as your services evolve.
The same reviews, on more of your code. Auto-discovered links feed directly into CodeRabbit's research agent, which explores linked repositories in real time and surfaces breaking changes with exact files and line numbers before you merge.
Auto-linked repositories are stored separately from manual links. Manual links are not overwritten. Automatic Repository Linking also builds on CodeRabbit’s knowledge base. If your organization has opted out of knowledge base features, both Multi-Repo Analysis and Automatic Repository Linking is disabled.
Cross-repo context has become a useful feature for coding agents and code reviewers, yet most approaches still depend on a human to draw the map. Some tools widen their view beyond a single repository only when you scope that context by hand, whether by passing extra directories to the agent, opening a multi-root workspace, or declaring related repositories in a configuration or agents file.
Others automate the discovery step but infer relationships from team activity, such as which contributors have recently committed where. That can show how people work, yet it does not prove that two repositories are actually related.
Distinctly, Automatic Repository Linking starts from code-level evidence. CodeRabbit looks across eligible repositories in your organization for signals such as imports, dependency manifests, API usage patterns, contracts, and repository README files. When those signals point to a real relationship, CodeRabbit links the repository as review context.
During review, CodeRabbit combines those auto-detected links with any repositories your team already configured manually. The multi-repo research agent can then inspect the linked repositories and surface downstream impact with concrete files and lines. The Review info section shows which repositories were considered.
You should not have to keep a manual map of every service, package, and repository relationship in your organization. With Automatic Repository Linking, CodeRabbit keeps that context closer to the code, so reviews can catch downstream impact before it reaches production.
Automatic Repository Linking is available for Pro plus and Enterprise customers. Enable it in your knowledge base settings and start sharing context between your repositories.
Learn more
