Skip to main content
POST
Create an automation

Authorizations

Authorization
string
header
required

Your Reason API key from Settings › API keys. New keys use reason_; legacy ara_ keys remain accepted. Keys are capability-scoped: run, mcp:read, mcp:write, secrets:read, secrets:write, sessions:read, sessions:debug, knowledge:read, memory:read, memory:write, skills:read, skills:write, repos:read, repos:write, reviews:read, reviews:write, deployment:read, analytics:read, org:read, org:write, attachments:read, attachments:write, guardrails:read, guardrails:write, network:read, network:write, automations:read, automations:write, agent_auth:read. network:read and network:write are opt-in, excluded from default keys. network:read reads the saved workspace network policy; network:write replaces custom allowed hosts and requires a current organization owner/admin. Legacy project PATCH requests containing network_allowed_hosts also require network:write plus their existing run scope and project management permission; saved project hosts do not affect current session egress. Only owners/admins can mint or manage keys with network:write; keys cannot grant themselves more authority. See /docs/network-policies for replacement semantics and runtime applicability. mcp:write manages MCP server configuration only; it does not authorize remote MCP-tool execution. sessions:read reads sessions, including the assistant, reasoning and tool activity in their events. sessions:debug is privileged: only for organization owners/admins, it expands session events to the full diagnostic projection (diagnostic event kinds, status text and raw event metadata).

Path Parameters

orgId
string
required

Organization id or slug. Resolve it with GET /v3/self.

Body

application/json

Creates the automation's configuration. It has no triggers until you add one via POST /automations/{automationId}/triggers, so it will not fire on its own yet.

instructions
string
required

The task each run carries out when a trigger fires, written as the prompt the agent receives.

Maximum string length: 100000
direct
object
name
string

Label that identifies the automation in lists and audit entries. Omit it to name the automation after the start of instructions.

Maximum string length: 200
repo
string

Connected GitHub repository, as owner/name, that each run checks out and works in; listRepositories lists the connected ones. Omit it for an automation whose runs start with no repository.

provider
enum<string>

Source-control host of repo, sent only together with it; on its own it returns 400 repo_required. Omit it to look the host up from the connected repositories.

Available options:
github
model
string | null

Model override. null uses the workspace default. New runs accept only the currently available Reason model catalog.

tools
object
enabled
boolean
default:true

Whether the automation starts active. Send false to create it paused: its triggers never fire until an update sets enabled: true.

Response

Created.

An automation: its instructions, tools, environment binding, and triggers.

id
string

Automation id; pass it as automationId to read, update, delete or add triggers.

org_id
string

Workspace the automation belongs to.

name
string | null

Label that identifies the automation.

instructions
string | null

The system instructions the automation runs with.

model
string | null

Model override for its runs; null uses the workspace default.

enabled
boolean

A paused (false) automation's triggers never fire.

tools
object

Tools the automation's runs may use.

environment
object

Repository each run starts in; repo is null for an automation without one.

triggers
object[]

Every condition that fires this automation. Empty until you add one.

warnings
object[]

Present only when the request was accepted but something is not as sent or not usable yet. Codes: schedule_interval_raised, app_tool_not_connected.

created_at
string<date-time>

When the automation was created.

updated_at
string<date-time>

When the automation's configuration last changed.