Pipeline properties and configuration

Pipeline properties refer to the configurable settings, variables, or inputs you use to customize the behavior of a pipeline. These parameters provide a means of making pipelines more adaptable, maintainable, and reusable by allowing you to configure their behavior without altering the core logic of the pipeline. When you add a new pipeline, you initialize its properties.You can also update the properties of an existing pipeline.

Pipeline parameters

To open the Pipeline Properties dialog, click . Note that you cannot update all fields. The pipeline properties dialog includes the following tabs:

Settings


Field	Description
Label	Required. The name for the pipeline. The pipeline name must be unique within the Project and must not exceed 100 characters.
Project	Required. The Project in which you save the Pipeline, displayed as <project_space>/<project_name>. All necessary resources (such as files and Accounts) must be within the same Project unless they are part of the Shared project. Note: You cannot update this property after you save your pipeline the first time. To change the Project later, you must move the pipeline to a new Project or make a copy.
Pipeline type	The default pipeline type is Standard. Agent and MCP Server types are set automatically when the corresponding Snaps are added to the pipeline canvas and cannot be manually set.
Tool	Enable this toggle to expose the pipeline as a tool that can be invoked by an Agent pipeline or an MCP Server. Full tool configuration — including Describe tool and Tool inputs — is managed in the Tool tab.
Resumable Mode	Select this checkbox to convert your pipeline into a resumable pipeline, which pauses when an error occurs and can be resumed at the point of error later.
Error Pipeline	The pipeline that handles any error documents produced by this Pipeline.
Error Pipeline Parameters	Key-value pairs containing parameters to pass to the Error Pipeline if an error occurs.
Parameters	Key-value pairs containing parameters to pass to this pipeline during execution. If the Capture checkbox is selected, the pipeline parameter value is displayed in the Pipeline executions in the SnapLogic Monitor, when the pipeline run is started by a Task or by another pipeline using the PipeExecute Snap. The keys and values of a pipeline parameter can contain alphanumeric characters only. You can also specify the values for the parameters within a Task that runs this pipeline. To use the value of the parameter as part of an expression within a Snap, you must prefix the parameter name with the underscore character ( _ ). For example, if the parameter name is `myParam1`, you can refer to its value by using `_myParam1` as follows: `''/out' + _myParam1 + '.json''`. To use the same account in multiple Pipelines, you can reference the account credentials using an expression in a Pipeline parameter value. Important: Use pipeline parameters to pass metadata, not data. To pass data, use any of the following methods: Save the data to a file and then pass the filename as a pipeline parameter value. Use the Pipeline Execute Snap and pass the data through the input view of the child pipeline. If you pass large data sets as parameter values, the following built-in limitations in the SnapLogic Platform can cause parameter values to be truncated or can prevent the pipeline from running: If Capture is selected, the first 256 bytes of the captured pipeline parameter value are stored in the runtime for debug purposes. Each Snaplex node pushes information to the control plane about all the Pipelines running on the JCC node, and this information is limited to 250 MB for all pipelines on the node. If the total size of the pipeline parameter values for one pipeline execution (including parameters that are not captured) exceeds 8 MB, the pipeline does not execute.
Expression Libraries	The list of expression libraries to import and use with this pipeline. Path: Path to the file that contains the expression to import. The path can be absolute or relative to the current project. As: The name to use for the library under the `lib` variable. Defaults to the base name of the file. Expression Libraries
Cache Pipelines	The data to pre-cache in a Cache Pipeline. The alias and corresponding pipeline to run and store the data under the provided alias. You can access this cache with Pipeline Cache properties, where you can call it from the cache and filter the conditions to return back the data specified by the function. This reference provides the full document from the match to then use the output from the target match.

Attention: You can use captured pipeline parameter values and Error Pipeline parameter values to search for a pipeline in the Pipeline executions in Monitor. For details, refer to Pipeline filter

Tool

The Tool tab is available on all pipelines. Use it to expose a pipeline as a tool that can be invoked by an Agent pipeline or an MCP Server, and to define the inputs the tool accepts.

The tab contains the following controls:


Control	Description
Expose this pipeline as a tool	Turn on this toggle to make the pipeline available to Agent Creator and the Multi Pipeline Function Generator Snap. When enabled, the Describe tool field and Tool inputs panel appear.
Describe tool	A plain-language description of what this pipeline does. The description is included in the tool definition sent to the LLM so the model can decide when to call this tool. Example: `gives weather forecast for a location using latitude and longitude`

Tool inputs

The Tool inputs panel defines the parameters the LLM must provide when calling this tool. These are separate from Pipeline Parameters (which use _paramName notation and are passed at pipeline execution time). Tool inputs use $paramName notation and are passed dynamically by the LLM at tool-call time.

When you first enable the Expose this pipeline as a tool toggle, choose one of the following editor modes. After saving, the mode is locked.


Editor mode	Description
Build with a form	Add inputs one at a time using a visual form. Each input row includes: Name — the parameter name (used as `$paramName` in expressions) Type — the data type (for example, String, Number, Boolean) Description — plain-language description for the LLM Required — whether the LLM must always provide this input To add inputs using pipeline expression suggestions, click + Add input — the form pre-populates suggestions based on detected `$paramName` references in the pipeline.
Use a JSON schema	Paste or upload a raw JSON Schema that defines the tool's input parameters. Use this mode when you have an existing schema or need to define complex types (nested objects, arrays, enums).
Start over	Discards all current tool input definitions and returns to the mode selection screen. The form and JSON schema editors are not interconvertible.

Note: Tool inputs defined here are stored separately from Pipeline Parameters and are not visible in the Open API tab. Existing tool pipelines that previously defined parameters in the Open API tab continue to work without changes.

Open API

You can use the Open API tab to add information, such as data type and description, about the Pipeline parameters. This information helps other Pipeline developers understand what each Pipeline parameter represents.

Pipeline parameters in this tab are populated after the pipeline is saved. If you added new parameters since the last time you saved the pipeline, save the pipeline again to update this tab.


Field	Description
Key	The name of the pipeline parameter as specified in the parameters section (Key).
Type	The data type of the parameter. Choose one of the following data types from the dropdown list: string number byte binary boolean date date-time password Default value: string https://swagger.io/docs/specification/data-models/data-types/ Data Types
Description	A description of the parameter. This description is included in the auto-generated API document about the pipeline’s properties. Default value: None
Required	Specifies whether the parameter is required. Default value: Not selected

Info


Field	Description
Notes	Additional information about this pipeline.
Purpose	The purpose of this pipeline.
Author	The creator of the pipeline. If you are adding a new pipeline, this field is prepopulated with your username or email address.
Doc Link	The link to the documentation about this pipeline (if any).