flagon-io/g1t

public

Where people and agents ship software together. The open-source git platform for the whole job: issues, agents, checks and deploys to the edge.

g1t.sh
Docs
Status

git agents cloudflare rust forge mcp

syntaqxThe sessions and plugin guides say a session is titled with what was typed (#52 ): since plugin 0.2.4, pasted text and image placeholders are left out of a session's title, a slash command or a prompt that is only a paste gives none so the next prompt titles the session, and the site reads a title that is only a tag as no title. The guides still said the first prompt that is not a slash command titles it. The plugin guide's hook table and Work with no issue yet now say so, with a title like <pasted_content id="5c50"> from an older plugin showing by its first prompt's intent or Session on its branch; the sessions guide's Live now and Work with no issue yet say the same.3d4f9d23d agoHistory

..
guest.rsA computer whose runner is registered again takes its own runner's name back instead of being refused. When the desktop app's sandbox was made again (a new VM, its registration gone), its runners registered under the names the old sandbox's still held on g1t, were refused with a 409 asking for --replace, and tried again forever. A runner now says which computer it is when it registers: a fingerprint g1t-runner works out from the operating system's machine id (Windows' MachineGuid, macOS's platform UUID, Linux's /etc/machine-id), hashed so the id never leaves the computer, or from --machine / G1T_RUNNER_MACHINE; a container sends none, since its machine id is its image's, and g1t-runner sandbox passes the computer's to the runner in the VM. The actions service keeps it hashed again with the workspace (migration 0022 adds runners.machine_hash), and a registration of a name already taken replaces that runner when it was registered from the same computer, or, for a runner from before fingerprints, when the same person made both tokens and the old runner has been offline for 90 seconds; a runner of the same name from another computer is still refused without --replace, and the refusal says so. The self-hosted runners guide covers registering again and --machine.5d ago
image.rsMerge the desktop runner's settings: a card for each kind of job, Linux and Windows, each with its state, job, cause and fix, and its own Start and Stop; an honest wait for a release's sandbox images, checked again by itself; and the same on the Runners page.5d ago
mod.rsGetting the runner running, a pass for quality (#50 ). What Windows says of itself is kept in the sandbox's folder until Windows restarts, since none of it changes before then: a status while something is missing no longer starts PowerShell every few seconds (2.8 s the first time, 1 ms after). Windows jobs with Windows Sandbox on but virtualization off in the firmware now say so before a job, not as it fails. Set up that turned a feature on and then couldn't install WSL still asks for the restart, and offers WSL again after it. Restart now is asked with Windows' own dialog, wherever the button is, so no page can restart a computer by itself.3d ago
sysbox.rsA computer whose runner is registered again takes its own runner's name back instead of being refused. When the desktop app's sandbox was made again (a new VM, its registration gone), its runners registered under the names the old sandbox's still held on g1t, were refused with a 409 asking for --replace, and tried again forever. A runner now says which computer it is when it registers: a fingerprint g1t-runner works out from the operating system's machine id (Windows' MachineGuid, macOS's platform UUID, Linux's /etc/machine-id), hashed so the id never leaves the computer, or from --machine / G1T_RUNNER_MACHINE; a container sends none, since its machine id is its image's, and g1t-runner sandbox passes the computer's to the runner in the VM. The actions service keeps it hashed again with the workspace (migration 0022 adds runners.machine_hash), and a registration of a name already taken replaces that runner when it was registered from the same computer, or, for a runner from before fingerprints, when the same person made both tokens and the old runner has been offline for 90 seconds; a runner of the same name from another computer is still refused without --replace, and the refusal says so. The self-hosted runners guide covers registering again and --machine.5d ago
vm.rsThe desktop app gets its runner running on a clean computer the way Docker Desktop does (#50 ). g1t-runner sandbox status names what a computer lacks for each kind of job, before anything is downloaded (needs, crates/runner/src/selfhosted/needs.rs): on Windows virtualization off in the firmware, a VM without nested virtualization, Virtual Machine Platform, WSL or an older WSL, Windows Home, Windows Sandbox, a Windows too old, a restart still to do; on Linux KVM, its group and QEMU; on macOS Homebrew, vfkit and macOS 13. Each is said only when Windows, /proc or the Mac says it for certain, so a computer that can run the sandbox is never held back; virtualization inside a VM is named from the error starting it gives. vfkit is found where Homebrew puts it, which a Mac app's PATH lacks.3d ago
wsl.rsGetting the runner running, a pass for quality (#50 ). What Windows says of itself is kept in the sandbox's folder until Windows restarts, since none of it changes before then: a status while something is missing no longer starts PowerShell every few seconds (2.8 s the first time, 1 ms after). Windows jobs with Windows Sandbox on but virtualization off in the firmware now say so before a job, not as it fails. Set up that turned a feature on and then couldn't install WSL still asks for the restart, and offers WSL again after it. Restart now is asked with Windows' own dialog, wherever the button is, so no page can restart a computer by itself.3d ago

..
guest.rs
image.rs
mod.rs
sysbox.rs
vm.rs
wsl.rs