41 lines
3.2 KiB
Markdown
41 lines
3.2 KiB
Markdown
# Workflow definitions
|
|
|
|
A workflow's definition is an object: an optional `description`, then `steps`, which run in order. The workflow's name is not in it. `save_workflow` takes it as its `definition` argument and `read_workflow` returns it, both as JSON. Unknown fields are refused.
|
|
|
|
Any workflow can run on a brief, each stage as its own task, and any workflow whose stages are single agents can be a project's review workflow (set with `set_review_workflow`), which reviews its tasks' snapshots: each stage is then one reviewer agent on its `model`, given the task's change and its stage's `prompt` (what to look for), and a stage after others also gets their reviews. A project can't review with a workflow that has a team stage (`agents`).
|
|
|
|
Each step is exactly one of:
|
|
|
|
| Step | Meaning |
|
|
|---|---|
|
|
| `stage` | A stage, by its name: one line, unique in the workflow. It runs as its own task, with `model` or `agents`, and `prompt`. |
|
|
| `order` | A list of steps run one after another. |
|
|
| `parallel` | A list of branches run at the same time, each a step (an `order` for a branch of several). |
|
|
|
|
A stage gets the outputs of the stages that end the step before it: the last stage of an `order`, or the last stage of every branch of a `parallel`. A stage with nothing before it starts from the brief.
|
|
|
|
| Stage field | Meaning |
|
|
|---|---|
|
|
| `model` | Also picks the vendor: `gpt-*` models run on Codex, `gemini-*` models on Antigravity, every other model on Claude Code. |
|
|
| `prompt` | The stage agent's instructions. |
|
|
| `agents` | In place of `model`: a list of at least two, each with `name`, `model` and `prompt`, that runs the stage as a team. They work in the stage's one task and worktree and talk through a shared board; each gets the stage's `prompt` and then its own, and the first agent finishes the stage. |
|
|
|
|
Where a stage's task branches from follows from the shape. A stage after exactly one stage starts from the commit that stage finished at. A stage after a `parallel` starts from the default branch and is told the branches it can merge. A stage with nothing before it starts where the workflow was started. A workflow is finished once every stage has finished; when it ends with a `parallel`, each branch's last stage keeps its task to merge, and the run's other stage tasks are discarded. A review is delivered to the task's agent, every stage's findings in one message, once every stage has reviewed. An `order` or `parallel` can't be empty.
|
|
|
|
```json
|
|
{
|
|
"description": "Build a change, review and test it in parallel, then fix what they found.",
|
|
"steps": [
|
|
{"stage": "Build", "model": "claude-opus-5-5", "prompt": "Implement the brief."},
|
|
{"parallel": [
|
|
{"stage": "Review", "model": "gpt-6.1-sol", "prompt": "Review the Build stage's work and list the problems."},
|
|
{"stage": "Test", "model": "gemini-4-argon-high", "prompt": "Exercise the change and report what fails."}
|
|
]},
|
|
{"stage": "Fix", "prompt": "Merge the branches you are given and address both reports.", "agents": [
|
|
{"name": "fixer", "model": "claude-opus-5-5", "prompt": "Fix every problem the reports list."},
|
|
{"name": "checker", "model": "gpt-6.1-sol", "prompt": "Check each fix as the fixer lands it."}
|
|
]}
|
|
]
|
|
}
|
|
```
|