BitBytecGet early access
Railway tracks branching toward a glowing signal tower, with robot-like and human-sized carriages travelling together.

TECH ·

Entire launches an agent-first GitHub rival that replaces pull requests

Entire, led by former GitHub chief executive Thomas Dohmke, says its new Git forge lets teams merge AI-written code using agent-scored checks.

News from

Picture: BitBytec · Original illustration, BitBytec

Entire, the start-up led by former GitHub chief executive Thomas Dohmke, has launched what it calls the first Git forge built for agents and humans to work together, a direct alternative to GitHub [1]. It was announced on 26 October and is available to every developer today, according to the company [1].

What changes

The headline change is Trails, which replaces the pull request, the standard way developers propose and review code. Instead of reading a diff, developers see a dashboard of evaluations scored by agents on every code push [1]. Results are available in Entire's web interface and its open-source command line tool, so coding agents can fix findings in a loop while a person checks progress [1]. When all the evaluations pass, Entire says code can be merged without looking at the code itself [1]. GeekWire adds that teams can still require a person's sign-off or passing automated tests before code goes in [2].

Dohmke argues that agents now write almost all code, faster and more complex than before [1]. GeekWire quotes his blog post saying the pull request "has become an audit log that only the agent reads" [2].

Speed, regions and memory

Entire says its hosting is built on a new Git-compatible server with pushes up to 89 times faster than GitHub and six times faster than Cursor Origin, based on its own open-source benchmarks [1][2]. Treat that as a company claim rather than independent testing. Private repositories can be hosted in the US, EU, India and Australia [1].

A third feature, Continuous Memory, syncs agent sessions into the Git repository so agents can search them. Entire says a codebase graph saves up to 71% on tokens, and an "agentic brain" turns past sessions into memories agents can learn from [1]. GeekWire reports the 71% figure came from the company's own testing [2].

Entire is also partnering with Buildkite and Depot for build and test pipelines, giving developers a route off GitHub Actions [1].

The wider picture

GeekWire reports the launch comes as GitHub strains under the load from AI agents, with several significant outages in recent months, including one in August that lasted nearly eight hours [2]. GitHub said developers and agents made 7.38 billion commits on its platform in September alone [2]. Cursor launched its own code-hosting service, Origin, in August [2].

Entire raised a $60 million seed round in February and now has 50 staff [1]. Its first releases mirrored existing GitHub repositories on a faster network; developers can still mirror from GitHub or now move their code across [1][2].

What it means for businesses

For non-technical leaders, the question behind this launch is who checks the work when software writes software. Entire's answer is automated evaluations, with human review optional. That could speed delivery, but it shifts trust to the quality of the checks, so the sensible approach is to keep human sign-off on anything critical. Public repositories and self-hosting are still on the roadmap [1], so larger organisations with strict hosting needs may need to wait. Nothing here proves it is better than GitHub; it shows that rivals now think code hosting needs rebuilding for agents.

Source: Entire — Entire Launches an Independent Git Forge for Agent-Human Collaboration

  • entire
  • github
  • git
  • ai agents
  • developer tools
  • code review

Sources

  1. Entire Launches an Independent Git Forge for Agent-Human Collaboration — Entire (primary source)
  2. Former GitHub CEO launches rival platform built for AI coding agents — GeekWire

Written with AI from the sources listed, checked against them before publishing.

Your idea. Your AI. Your creation.

What would you make first?

Describe it to the chat you already use. Accept the plan on your Mac. Watch your own coding agent build it.