Static Constants
Group fixed values (model, thresholds, API keys) in one node — each constant exposes its own output port
What does the Static Constants node do?
The Static Constants node groups several fixed values you reuse across the workflow. Each constant becomes an output port you wire separately — handy when the same setting feeds multiple nodes.
Three typical use cases:
- Shared settings — e.g.
model=claude-sonnet-4-6andtemperature=0.2, wired to several LLM nodes. Change the value once. - Common thresholds — e.g.
min_score=0.7, wired to several Conditional nodes to apply the same rule everywhere. - Workflow API key — e.g.
api_tokenwith Encrypt this value enabled: the value is masked (••••••••) and decrypted only at execution time.
Static Text vs Static Constants. Static Text = one fixed string. Static Constants = several named values, each with its own port (plus optional encryption for secrets).
Quick setup
Add the node to the canvas
Node Library → Inputs → Static → Static Constants.
Add your constants
Click + Add constant for each value:
- Name — port identifier (e.g.
model,api_token). Letters, digits, and_only — no spaces. - Value — the value emitted at execution time.
- Encrypt this value — enable for secrets (tokens, API keys).
Connect the outputs
Each constant appears as a port on the right of the node (model, api_token, …). Connect each port to the node that needs it.
Configuration parameters
Required fields
Name string required default: Static Constants Node name — Label on the canvas (e.g. “LLM settings”, “API tokens”).
Description string required default: Define named constants (plaintext or encrypted) and expose them as separate outputs Node description — Short phrase describing what these constants are for.
Constants list required List of constants — Each row has a Name, a Value, and the Encrypt this value toggle. Names must be unique. An encrypted value cannot be left empty.
Editing an encrypted secret — the value is not shown in plain text in the UI. Click Replace to enter a new value; it will be re-encrypted on save.
What does the node output?
One port per constant, each emitting a string.
<constant_name> string The value defined in the settings. For an encrypted constant, the platform decrypts it at execution time — it never appears in plain text in the database.
Example: model → claude-sonnet-4-6, api_token → decrypted token value.
Examples
Centralised LLM settings
| Name | Value | Encrypted |
|---|---|---|
model | claude-sonnet-4-6 | no |
temperature | 0.2 | no |
Wire model and temperature to your LLM nodes.
Secured API token
| Name | Value | Encrypted |
|---|---|---|
service_base_url | https://internal.dng.ai/v2 | no |
service_token | (your token) | yes |
Wire both to an HTTP Request node — the token is decrypted when the call is made.
Best practices
Separate secrets from settings. One node for encrypted tokens, another for model / temperature / thresholds — easier to read and manage.
Encryption protects storage, not edit rights. Anyone who can edit the workflow can replace an encrypted value (without reading it). For enterprise secrets, prefer the platform’s OAuth integrations.
Common issues
'Constant name is invalid' error
Cause: the name contains a space, dash, or starts with a digit.
Solution: use api_token, min_score, MODEL_NAME.
'Duplicate constant name' error
Cause: two constants share the same name.
Solution: rename one of them.
'Encrypted value is required' error
Cause: encryption is enabled but the Value field is empty.
Solution: enter a value or disable encryption.
Output ports don't update
Cause: changes were not saved.
Solution: click Update tool in the settings panel.