Execution Data
Execution Data
Section titled “Execution Data”Read the current run’s context, or attach searchable metadata to it.
What it does
Section titled “What it does”Every run is otherwise anonymous: run history can only show a timestamp and a status, and a node has no way to know which run it is part of. This node closes both gaps, in two halves that are useful on their own:
- Read run context — outputs this run’s id, the trigger type that started it, and its timing, so a later node can branch on how the run began or how long it has been going, and so a log line can name the run.
- Attach metadata — writes a searchable key/value pair onto the run’s record, turning “the run at 14:03” into “the run that processed order 4471”. The pair is saved with the run and is searchable in run history.
When to use it
Section titled “When to use it”- Make run history findable by something meaningful — an order number, a customer id, a batch name — instead of only by time and status. Attach the value with Attach metadata, then search for it in the run list.
- Branch on how a run started. Read the trigger type and route a scheduled run differently from a manual one, or a chat-started run differently from a page-load one.
- Annotate logs with the run id so a support conversation can point at one exact run.
Inputs and settings
Section titled “Inputs and settings”| Setting | Notes |
|---|---|
| Operation | Read run context (output the run’s id, trigger type and timing) or Attach metadata (save a key/value pair on the run). Defaults to Read run context. |
| Key | The name of the metadata field to attach — for example orderId. Required for Attach metadata. |
| Value | The value to store under the key — for example 4471. Non-text values are stored as text so they stay searchable. |
The trigger type is read from the trigger node upstream of this one, the same information that flows down every branch of a run. If nothing upstream identifies the trigger, the trigger type reads as unknown.
Outputs
Section titled “Outputs”Read run context outputs an object:
| Field | Meaning |
|---|---|
runId | Stable id for this run, or null outside a workflow run. |
workflowId | The workflow this run belongs to, or null. |
triggerType | How the run started — schedule, manual, elementChange, chat, and so on — or unknown. |
startedAt | When the run started, as an ISO timestamp, or null. |
elapsedMs | Milliseconds since the run started, or null. |
Attach metadata outputs the pair it wrote plus the run’s full metadata map so far:
| Field | Meaning |
|---|---|
key / value | The pair that was written (value coerced to text). |
metadata | Every key/value attached to this run so far. |
persisted | true when the pair was attached to a real run; false when the node ran on its own (a single-node test), where there is no run to annotate. |
Where metadata shows up
Section titled “Where metadata shows up”Attached metadata appears in run history: each run’s row shows its metadata, the detail panel lists every pair and the run id, and the history search box matches a run by its metadata (either half of a pair), its run id, or its status. Search “4471” and the run that processed order 4471 comes back.
Dependencies and credentials
Section titled “Dependencies and credentials”- No credential or node dependency is required. The node runs entirely in the browser and works without an account.
- Metadata is saved with runs recorded in this browser profile (local mode). Run context — id, trigger type, timing — is available in every mode.
- Metadata is attached to this run, and runs already belong to a single workflow, so there is no shared store an imported workflow could read another workflow’s metadata out of. The epic’s per-workflow scoping holds automatically.
Troubleshooting
Section titled “Troubleshooting”persisted: falseon a set means the node ran outside a workflow run (a single-node test). Run the whole workflow and the pair is attached to that run.triggerTypeisunknownwhen no upstream trigger labels itself. Put the Execution Data node on a branch downstream of the trigger so the trigger type reaches it.- Metadata is text. A number or boolean flowing into Value is stored as its text form so it stays searchable; read it back as a string.
- A resumed run is a new run. A run that paused and resumed is recorded as its own execution with its own id; metadata attached before it paused is not carried across the pause.
Related nodes
Section titled “Related nodes”- Data Store — remembers values between runs; Execution Data annotates this run
- Digest — accumulate items across runs and release them together
- Node types overview