Overview - Entire

What are Entire Repositories?

Entire Repositories are Git repositories stored in EntireDB, Entire’s distributed Git database. They store source code, branches, commits, and the agent context connected to that work. That means a repository can contain more than the final code change: it can also point to the sessions, checkpoints, transcripts, and review history that explain how the code got there. To get an existing GitHub repository onto Entire’s network, create a repository mirror, a synced, read-optimized copy served from the region you choose. Teams can keep working in GitHub, while developers and coding agents can use the regional Entire mirror as a Git remote. This gives high-volume clone and fetch traffic a Git remote in the selected Entire region instead of sending every read back through GitHub.

How do they work?

EntireDB is built around two kinds of Git data:

• repository data that never changes, such as Git objects
• repository state that must stay consistent, such as branches and refs

An Entire Repository lives in the region you choose. EntireDB replicates its Git data within that region for resiliency, and coordinates branch and ref updates there so readers see the current repository state. Supported regions are in the United States and the European Union. This lets many agents clone, fetch, inspect, and search the same repository at the same time without routing every read through one central Git host. Reads are served from the Entire region you choose, while branch and ref updates stay coordinated in that region.

Why do they matter?

Agent-driven development changes the shape of repository traffic. A developer might clone a repository once, pull changes throughout the day, and push when work is ready. Agents can create a much higher-frequency loop: clone repositories, inspect files, search history, replay context, and run repeated checks. As teams add more agents, that traffic can add latency, increase the impact of central Git host outages, and run into rate limits on the source repository. Entire Repositories are designed for that workload:

• Regional control: store and serve repository data from the Entire region your team chooses.
• Consistency: keep changing branch state coordinated as humans and agents push work.
• Scale: support many concurrent agent reads without overloading a single Git host.
• Sovereignty: keep sensitive code in the region your team chooses.
• Traceability: keep agent sessions and checkpoints connected to the repository they changed.

Continue with

[**GitHub App Access**
Choose which GitHub repositories Entire can access.](https://docs.entire.io/guides/repositories/github-app-access)

[**Mirrors**
Bring an existing GitHub repository into Entire.](https://docs.entire.io/guides/repositories/mirrors)

[**Mirrors in CLI**
Create a mirror from the terminal and use it from Git.](https://docs.entire.io/guides/repositories/mirrors-in-cli)