Operations

Operations are the individual steps inside a flow. CairnCMS ships with a built-in set of operations covering data work, branching, communication, scripting, and HTTP. Custom operations can be added through the extension system.

Every operation appends a value to the data chain under its own key. Operations that do not produce data still append null so that downstream operations can reference them predictably.

Condition

Routes execution to the success or failure path based on a filter rule.

• Condition Rules — a filter rule evaluated against the data chain

Use this when a flow needs to branch on data the trigger or earlier operations produced. The condition does not generate data of its own; it appends null regardless of which path is taken.

If the filter rule itself is misconfigured, the operation appends a debug array describing the misconfiguration.

When used at the end of a flow on a Filter (blocking) Event hook trigger, a failure-path Condition cancels the original database transaction.

Run Script

Runs custom JavaScript inside an isolated V8 sandbox.

The sandbox is fully isolated from the host: no file system access, network access, or require() of built-in or external modules. The script receives the data chain as its argument and returns a JSON-serializable value, which is appended under the operation key. A console shim is available and routes to the platform logger.

The script’s process.env is populated from the operator-allow-listed environment variables (governed separately by FLOWS_ENV_ALLOW_LIST).

Throwing inside a script ends the flow. On a Filter (blocking) Event hook trigger, throwing also cancels the original event transaction.

The isolate has a configurable memory and time budget, set via FLOWS_RUN_SCRIPT_MAX_MEMORY (default 32, MB) and FLOWS_RUN_SCRIPT_TIMEOUT (default 10000, ms). Scripts that exceed either limit are aborted.

Create Data

Creates one or more items in a collection.

• Collection — the target collection
• Permissions — the role whose permissions the operation runs under
• Emit Events — whether to fire event hooks for this create
• Payload — the data to insert

Appends an array of created item IDs.

Read Data

Reads items from a collection by ID or query.

• Collection — the source collection
• IDs — specific item IDs to fetch
• Query — a filter rule for selecting items
• Permissions — the role whose permissions the operation runs under
• Emit Events — whether to fire event hooks

Appends the result of the read. With a single ID, this is a single item; with multiple IDs or a query, this is an array of items.

Update Data

Updates one or more items in a collection by ID or query.

• Collection — the target collection
• IDs — specific item IDs to update
• Query — a filter rule for selecting items to update
• Payload — the partial update to apply
• Permissions — the role whose permissions the operation runs under
• Emit Events — whether to fire event hooks

Appends the updated key(s). With a single ID, this is a single primary key; with multiple IDs or a query, this is an array of primary keys.

Delete Data

Deletes items from a collection by ID or query.

• Collection — the target collection
• IDs — specific item IDs to delete
• Query — a filter rule for selecting items to delete
• Permissions — the role whose permissions the operation runs under
• Emit Events — whether to fire event hooks

Appends the deleted key(s). With a single ID, this is a single primary key; with multiple IDs or a query, this is an array of primary keys.

For all four data operations, Emit Events controls whether the operation fires its own event hooks. Turn it off when this operation might trip the same flow that contains it, to prevent infinite loops.

Send Email

Sends one or more emails through the configured email provider.

• To — one or more email addresses
• Subject — the email subject
• Type — markdown, wysiwyg, or template
• Body — the email body (when Type is markdown or wysiwyg)
• Template — the name of an installed email template (when Type is template; defaults to base)
• Data — variables to pass into the template (when Type is template)

Appends null.

Send Notification

Sends an in-app notification to one or more users.

• Users — one or more user UUIDs
• Title — the notification title
• Message — the notification body
• Permissions — the role whose permissions the operation runs under

Appends an array of notification IDs.

Webhook / Request URL

Sends an HTTP request to an arbitrary URL.

• Method — GET, POST, PATCH, DELETE, or others
• URL — the target URL
• Headers — request headers
• Request Body — the body to send

Appends the response under the operation key.

For richer outgoing webhook patterns, see the Webhooks page.

Trigger Flow

Runs another flow and optionally passes data to it. The other flow must use the Another flow trigger.

• Flow — the UUID of the flow to trigger
• Payload — the JSON to pass to the other flow’s $trigger
• Iteration Mode — when the payload is an array: parallel (default; runs all in parallel), serial (runs one after another), or batch (runs in batches)
• Batch Size — the batch size when Iteration Mode is batch (defaults to 10)

Appends the other flow’s response, when one is configured. If the other flow has no Response Body, this operation appends null.

If the payload is an array, the other flow runs once per array item using the configured iteration mode.

Sleep

Pauses the flow for a configured number of milliseconds.

• Milliseconds — how long to sleep

Appends null. Use sparingly; long sleeps tie up flow workers.

Transform Payload

Builds a custom JSON object that subsequent operations can reference.

• JSON — the object to build, with data chain variables for dynamic values

Appends the resulting object.

This operation is useful for stitching together values from earlier operations into a single shape, such as preparing the body of a webhook request or the payload for a Create Data operation.

Log to Console

Writes a message to the server-side console and to the flow’s logs.

• Message — the message to log

Appends null. The message itself is visible only in the flow’s logs and in the server output, not in the data chain.

This is the primary debugging tool for flows.