Pick any line to see why it is the way it is: the commit, the pull request and issue it came from, and what the agent was thinking.
| Agents and memory, checks and conflicts, profiles, slug renames, custom domains | 1 | import type { AgentRun } from "./agents"; |
| A free allowance on g1t's models, so anyone can try its agents | 2 | import type { Trial } from "./billing"; |
| Hosted agents: sandboxes on Cloudflare Containers started from an intent | 3 | import type { User, Viewer } from "./identity"; |
| 4 | import type { Result } from "./result"; | |
| Issues and pull requests replace intents and attempts | 5 | import type { RepoPath } from "./repos"; |
| Thirteen MCP tools and classic token scopes; agents rate their confidence and can be put on an issue in one step | 6 | import type { DelegateInput, Delegated, Plan, Pull } from "./work"; |
| g1t agents: model menu and optional AI Gateway routing | 7 | |
| Hosted agents: sandboxes on Cloudflare Containers started from an intent | 8 | export type RunHostedInput = { |
| Agents as a team: lifecycle, merge queue, billing and a new shell | 9 | /** Extra guidance given to the agent along with the issue. */ |
| Hosted agents: sandboxes on Cloudflare Containers started from an intent | 10 | instructions?: string; |
| 11 | }; | |
| 12 | ||
| Agents as a team: lifecycle, merge queue, billing and a new shell | 13 | /** |
| Fast pages, required checks on the branch, self-hosted runners, honest incidents | 14 | * Sandboxes on g1t: agents that work on an issue, reviews, and the merge queue. |
| Agents as a team: lifecycle, merge queue, billing and a new shell | 15 | * |
| 16 | * Nobody who assigns a g1t agent picks a model. g1t routes each kind of | |
| 17 | * work itself, and says in the session which model ran. | |
| 18 | */ | |
| Models per workspace: several providers, routed by kind of work | 19 | /** |
| 20 | * How a workspace's agents reach a model. The workspace decides: its own | |
| 21 | * provider (`own` names it), or g1t's hosted models paid from its credit. | |
| 22 | */ | |
| 23 | export type ModelAccess = { | |
| 24 | /** The workspace's own model connection, by name, if it has one. */ | |
| 25 | own: string | null; | |
| A free allowance on g1t's models, so anyone can try its agents | 26 | /** Whether g1t's hosted models are open to it, on its free allowance or otherwise. */ |
| Models per workspace: several providers, routed by kind of work | 27 | hosted: boolean; |
| A free allowance on g1t's models, so anyone can try its agents | 28 | /** Its free allowance, when that is how it reaches g1t's hosted models; null when it needs none. */ |
| 29 | trial: Trial | null; | |
| Git storage hardened, pages in tens of milliseconds, honest security alerts, and costs reconciled daily | 30 | /** |
| 31 | * True when hosted models are closed to it because billing takes no real | |
| 32 | * money yet: until then they are open only to the workspaces g1t lists, | |
| 33 | * and every other workspace uses its own provider. | |
| 34 | */ | |
| 35 | preview: boolean; | |
| Models per workspace: several providers, routed by kind of work | 36 | }; |
| 37 | ||
| Agents get guardrails, run credentials, an audit log, a context hub, repository instructions and mentions; security upkeep; snake_case API | 38 | /** |
| 39 | * The files every g1t agent run in a repository reads as the repository's | |
| 40 | * instructions, as they are on its default branch. | |
| 41 | */ | |
| 42 | export type RepoInstructions = { | |
| 43 | branch: string; | |
| 44 | /** Null when the repository has no commits yet. */ | |
| 45 | commit: string | null; | |
| 46 | files: { | |
| 47 | path: string; | |
| 48 | /** `root` is read by every run, `directory` by runs that touch files under it, `review` by reviews. */ | |
| 49 | role: "root" | "directory" | "review"; | |
| 50 | text: string; | |
| 51 | /** Longer than agents are given; they get the start of it. */ | |
| 52 | truncated: boolean; | |
| 53 | lastChanged: { commit: string; message: string; author: string; at: string } | null; | |
| 54 | }[]; | |
| 55 | /** How much of each file, and of all of them, an agent is given. */ | |
| 56 | limits: { fileChars: number; totalChars: number }; | |
| 57 | }; | |
| 58 | ||
| Hosted agents: sandboxes on Cloudflare Containers started from an intent | 59 | export interface RunnerApi { |
| Agents get guardrails, run credentials, an audit log, a context hub, repository instructions and mentions; security upkeep; snake_case API | 60 | /** |
| 61 | * The repository's instructions for agents (`AGENTS.md`, `CLAUDE.md`, | |
| 62 | * `.g1t/review.md`), as they are on its default branch: what every g1t | |
| 63 | * agent run there reads. Whoever can see the repository may ask. | |
| 64 | */ | |
| 65 | instructions(viewer: Viewer, repo: RepoPath): Promise<Result<RepoInstructions>>; | |
| Models per workspace: several providers, routed by kind of work | 66 | /** How `workspace`'s agents would reach a model now. */ |
| 67 | modelAccess(workspace: string): Promise<ModelAccess>; | |
| Hosted agents: sandboxes on Cloudflare Containers started from an intent | 68 | /** |
| g1t's agents only for listed workspaces, whatever the state of billing | 69 | * Whether `viewer` may put g1t's agents to work: in `repo`'s workspace, |
| 70 | * or with none named, in any of theirs. | |
| 71 | */ | |
| 72 | enabled(viewer: Viewer, repo?: RepoPath): Promise<boolean>; | |
| 73 | /** | |
| Agents as a team: lifecycle, merge queue, billing and a new shell | 74 | * Assigns the issue to a g1t agent: opens a draft pull request for it, |
| 75 | * made by an agent in a sandbox of its own. Returns as soon as the | |
| 76 | * sandbox is starting; progress shows up in the pull request's session. | |
| 77 | * Scale comes from assigning many issues, each to its own agent. | |
| Hosted agents: sandboxes on Cloudflare Containers started from an intent | 78 | */ |
| Agents as a team: lifecycle, merge queue, billing and a new shell | 79 | run(actor: User, repo: RepoPath, issue: number, input?: RunHostedInput): Promise<Result<Pull>>; |
| Acceptance checks in sandboxes, line comments and review verdicts | 80 | /** |
| Thirteen MCP tools and classic token scopes; agents rate their confidence and can be put on an issue in one step | 81 | * Puts an agent on something in one step: opens an issue and assigns it |
| g1t is one name: its agent's work, commits and comments show as @g1t, and nobody can claim g1t or g1t-agent | 82 | * to g1t. Refused, with nothing opened, unless `actor` may put |
| Thirteen MCP tools and classic token scopes; agents rate their confidence and can be put on an issue in one step | 83 | * agents to work in `repo` (Write). Once opened, the issue stays whatever |
| 84 | * becomes of the agent: `agent` says whether it started, waits for a | |
| 85 | * free slot, or did not start, why and where to fix it. | |
| 86 | */ | |
| 87 | delegate(actor: User, repo: RepoPath, input: DelegateInput): Promise<Result<Delegated>>; | |
| 88 | /** | |
| Agents as a team: lifecycle, merge queue, billing and a new shell | 89 | * Has an agent read the repository and turn an outcome into a plan: the |
| 90 | * issues that would get there and the order they have to land in. Returns | |
| 91 | * the plan's id as soon as the sandbox is starting; the plan fills in when | |
| 92 | * the agent has written it. Members of the repository's workspace only. | |
| 93 | */ | |
| 94 | plan(actor: User, repo: RepoPath, brief: string): Promise<Result<{ planId: string }>>; | |
| 95 | /** | |
| 96 | * Opens a plan's issues. With `assign`, a g1t agent starts on each that | |
| 97 | * depends on nothing, and on the others as what they depend on merges. | |
| 98 | */ | |
| 99 | applyPlan( | |
| 100 | actor: User, | |
| 101 | repo: RepoPath, | |
| 102 | planId: string, | |
| 103 | options?: { assign?: boolean; keep?: number[] }, | |
| 104 | ): Promise<Result<Plan>>; | |
| 105 | /** | |
| 106 | * Brings a pull request up to date with the branch it would merge into, | |
| 107 | * in a sandbox. A clean merge is pushed as it is; a conflict is resolved | |
| 108 | * by a g1t agent. Whoever can push to the pull request's source may ask: | |
| 109 | * its author, or for one from a branch, a workspace member. | |
| 110 | */ | |
| 111 | update(actor: User, repo: RepoPath, number: number): Promise<Result<boolean>>; | |
| 112 | /** | |
| 113 | * Has a g1t agent review a pull request: line comments, a summary and a | |
| g1t is one name: its agent's work, commits and comments show as @g1t, and nobody can claim g1t or g1t-agent | 114 | * verdict, posted as `g1t`. |
| Agents as a team: lifecycle, merge queue, billing and a new shell | 115 | */ |
| 116 | review(actor: User, repo: RepoPath, number: number): Promise<Result<boolean>>; | |
| Agents and memory, checks and conflicts, profiles, slug renames, custom domains | 117 | /** |
| 118 | * Stops an agent run: marks it stopped, destroys its sandbox, and leaves | |
| 119 | * the pull request it was on for a person. Members of the workspace only. | |
| 120 | */ | |
| 121 | stopRun(actor: User, repo: RepoPath, runId: string): Promise<Result<AgentRun>>; | |
| Hosted agents: sandboxes on Cloudflare Containers started from an intent | 122 | } |