Skip to content
⌘K
Explore
Download
Docs
Sign in
Sign up
flagon-io
/
g1t
public
Star
0
Overview
Code
Issues
Pull requests
Workflows
Deployments
Insights
Files
Commits
Branches
Tags
Releases
Compare
c28e1acc57656a00d74f5831facb26aad88db175
g1t
/
…
/
crates
/
runner
/
src
Claude
agent
Merge branch 'worktree-agent-af75396ad4c6ee7a8' into claude/cool-newton-7nm8r9
c28e1ac
yesterday
History
..
actions
docker
selfhosted
abuse.rs
backup.rs
bump.rs
checks.rs
clone.rs
computer.rs
confidence.rs
deploy.rs
guard.rs
harness.rs
learned.rs
main.rs
mergecheck.rs
plan.rs
progress.rs
queue.rs
reply.rs
report.rs
review.rs
revise.rs
steer.rs
update.rs
..
actions
What runs differently on g1t's machines is said before it surprises anyone, and four things now behave as on GitHub. A commit whose message says `[skip ci]`, `[ci skip]`, `[no ci]`, `[skip actions]` or `[actions skip]`, or ends with `skip-checks: true`, starts no `push` or `pull_request` workflows. A `pull_request_review` event's `review.state` is `approved` or `changes_requested`, not `approve` or `requestchanges`. A self-hosted job that names no `timeout-minutes` runs for 360 minutes, not 60. A JavaScript action's `::debug::` lines show on a debug run, as `run:` steps' do. The workflow's page now warns when a job asks for more than the 60 minutes g1t's machines give it, when an Arm label will run on x64, and when a matrix on a job that calls a reusable workflow will run it once; it no longer says artifacts are kept up to 60 MB (they are kept up to 5 GiB), nor that `setup-node` with `cache:` runs without its cache (the toolkit's cache works). The desktop release's two hosted jobs ask for the 60 minutes they get. The Actions and self-hosted runners guides say so.
yesterday
docker
Merge branch 'main' into actions-toolkit-oidc-artifacts
5d ago
selfhosted
Merge branch 'main' into actions-toolkit-oidc-artifacts
5d ago
abuse.rs
Invite-only launch: sign in with GitHub, repository access and lifecycle, many emails, a new look
1w ago
backup.rs
Merge remote-tracking branch 'origin/main' into workspace-chat
4d ago
bump.rs
Teams and CODEOWNERS, labels and milestones, dependency updates, the security suite, and a clearer top bar
5d ago
checks.rs
Fast pages, required checks on the branch, self-hosted runners, honest incidents
1w ago
clone.rs
The docs folder is gone, and what it held lives where people read it: how a self-hosted g1t runs and how to deploy g1t to Cloudflare are pages on docs.g1t.sh under Run g1t yourself, and speed, rate limits and operating g1t.sh are sections of CONTRIBUTING.md; code that cited a file in docs/ now points to the page or section that covers it, or says what it means itself, and applied migrations and the runner images are left as they were.
3d ago
computer.rs
Every agent can have its own computer. A session that needs one wakes it: a home of its own on g1t cloud, one per agent and never shared, where it runs commands, reads and writes files and keeps what it made, with each session working in its own folder under a shared home; after ten idle minutes it sleeps, its home kept as a snapshot and restored when it wakes, and Reset wipes the home while memory and artifacts stay. Its shell and files are abilities with the usual choices, Alone, Alone when asked, Ask first or Never, offered only inside sessions and never to a chat reply; every command shows on the session with its output, and the agent's new Computer tab shows the state, the disk used of the five gigabytes included, the recent commands, and Wake, Put to sleep and Reset. Machine time counts only while it is awake, on the sandbox lines of the ledger that name the agent and who asked, held to the same spend caps as the session; the disk itself costs nothing in this version. The runner gained a long-lived supervisor that answers the computer's requests inside the container, and the runner service a computer per agent that keeps its snapshot in the agent homes bucket when one is attached, and says so when none is. The REST API and the agent tool can read a computer, wake it, put it to sleep and reset it. The agents, abilities, sessions, runners, billing and deploy guides say how it works and what an operator sets up; pinning a computer to your own runner, its browser and take-over come next.
2d ago
confidence.rs
Agents no longer count not having run CI as a doubt: the workflows run as checks after the push
6d ago
deploy.rs
Previews build from any branch; Agent in place of Ask AI; pin from the sidebar
5d ago
guard.rs
Reviews, plans and replies are written where the guardrail allows: g1t's answer files are inside it
6d ago
harness.rs
Merge branch 'model-routing'
5d ago
learned.rs
Agents get guardrails, run credentials, an audit log, a context hub, repository instructions and mentions; security upkeep; snake_case API
1w ago
main.rs
Every agent can have its own computer. A session that needs one wakes it: a home of its own on g1t cloud, one per agent and never shared, where it runs commands, reads and writes files and keeps what it made, with each session working in its own folder under a shared home; after ten idle minutes it sleeps, its home kept as a snapshot and restored when it wakes, and Reset wipes the home while memory and artifacts stay. Its shell and files are abilities with the usual choices, Alone, Alone when asked, Ask first or Never, offered only inside sessions and never to a chat reply; every command shows on the session with its output, and the agent's new Computer tab shows the state, the disk used of the five gigabytes included, the recent commands, and Wake, Put to sleep and Reset. Machine time counts only while it is awake, on the sandbox lines of the ledger that name the agent and who asked, held to the same spend caps as the session; the disk itself costs nothing in this version. The runner gained a long-lived supervisor that answers the computer's requests inside the container, and the runner service a computer per agent that keeps its snapshot in the agent homes bucket when one is attached, and says so when none is. The REST API and the agent tool can read a computer, wake it, put it to sleep and reset it. The agents, abilities, sessions, runners, billing and deploy guides say how it works and what an operator sets up; pinning a computer to your own runner, its browser and take-over come next.
2d ago
mergecheck.rs
g1t is one name: its agent's work, commits and comments show as @g1t, and nobody can claim g1t or g1t-agent
1w ago
plan.rs
Reviews, plans and replies are written where the guardrail allows: g1t's answer files are inside it
6d ago
progress.rs
Invite-only launch: sign in with GitHub, repository access and lifecycle, many emails, a new look
1w ago
queue.rs
g1t is one name: its agent's work, commits and comments show as @g1t, and nobody can claim g1t or g1t-agent
1w ago
reply.rs
Reviews, plans and replies are written where the guardrail allows: g1t's answer files are inside it
6d ago
report.rs
Agents as a team: lifecycle, merge queue, billing and a new shell
1w ago
review.rs
Reviews, plans and replies are written where the guardrail allows: g1t's answer files are inside it
6d ago
revise.rs
Fast pages, required checks on the branch, self-hosted runners, honest incidents
1w ago
steer.rs
Agents get guardrails, run credentials, an audit log, a context hub, repository instructions and mentions; security upkeep; snake_case API
1w ago
update.rs
g1t is one name: its agent's work, commits and comments show as @g1t, and nobody can claim g1t or g1t-agent
1w ago