Pipelines are saved multi-step workflows built from nodes — remove background, upscale, generate, and more — chained together in a graph. Create pipelines in the Pixa web app, then run them from any MCP-compatible agent.Documentation Index
Fetch the complete documentation index at: https://pixa.com/docs/llms.txt
Use this file to discover all available pages before exploring further.
Running a pipeline
Use thepipelines tool with action run:
pipelines (action: get) to inspect the pipeline, then pipelines (action: run) with inputs mapped to the correct node IDs.
Inputs are mapped by node alias. Values can be public URLs, Pixa asset IDs, or text — the agent picks the right type based on the node’s input spec.
MCP actions
Thepipelines tool supports these actions:
| Action | Description |
|---|---|
list | List all pipelines |
search | Find pipelines by name |
get | Get pipeline details with nodes and edges |
run | Start a pipeline execution |
list_runs | List runs for a specific pipeline |
get_run | Get run details with per-node status |
cancel_run | Stop a running execution |
list_node_types | Available node types with I/O specs |
display_pipeline to render an interactive node graph preview in MCP App UIs.
Finding pipelines
pipelines (action: search) with the relevant query, or list to browse everything available.
Monitoring runs
Pipeline runs are async. Thepipelines (action: run) call returns a run_id. Track progress with:
pipelines (action: get_run) to fetch the run state, including per-node status and outputs.
End-to-end example
- Calls
pipelines(search) to find the “Product Cleanup” pipeline. - Calls
pipelines(get) to inspect the input nodes. - Calls
pipelines(run) with the photo mapped to the right input andcollection_idset to the target collection. - Polls
pipelines(get_run) until the run completes, then surfaces the resulting assets.
For more on pipeline concepts and the web-based pipeline editor, see the Pipelines user guide.