Pipelines

Pipelines chain operations into multi-step DAG workflows with typed nodes and edges.

Structure

Field	Description
id	Unique identifier
name	Human-readable name
description	What this pipeline does
nodes	Array of pipeline nodes
edges	Array of edges connecting nodes
tags	Categorization labels

Node Types

Folder Node

Provides a directory tree as input to downstream nodes.

{
  "id": "folder-1",
  "type": "folder",
  "data": {
    "label": "Source Code",
    "folderPath": "/path/to/project"
  }
}

Code File Node

Provides a single file as input.

{
  "id": "file-1",
  "type": "code-file",
  "data": {
    "label": "README",
    "filePath": "README.md"
  }
}

Operation Node

Executes a registered operation.

{
  "id": "op-1",
  "type": "operation",
  "data": {
    "label": "Lint Check",
    "operationId": "lint-check"
  }
}

Output Local Path Node

Writes pipeline output to a local directory.

{
  "id": "output-1",
  "type": "output-local-path",
  "data": {
    "label": "Results",
    "localPath": ".ordine/results"
  }
}

Edges

Edges define data flow between nodes:

{
  "id": "e1",
  "source": "file-1",
  "target": "op-1"
}

Execution Model

Level 0:  [folder-1]  [file-1]     ← Input nodes (parallel)
              │            │
Level 1:  [op-check]  [op-lint]    ← Operations (parallel)
              │            │
Level 2:       [op-merge]           ← Merge step
                   │
Level 3:      [output-1]            ← Output

1. The pipeline engine performs a topological sort on the DAG
2. Nodes are grouped into execution levels
3. Nodes at the same level run in parallel
4. Each node receives input from its parent nodes via NodeCtx
5. Output flows to child nodes along edges

Creating a Pipeline

Via the REST API

curl -X PUT http://localhost:9433/api/pipelines/my-pipeline \
  -H 'Content-Type: application/json' \
  -d '{
    "name": "Code Review Pipeline",
    "description": "Automated code review with linting and AI analysis",
    "nodes": [...],
    "edges": [...],
    "tags": ["review"]
  }'

Via the Canvas UI

1. Navigate to Pipelines and create a new pipeline
2. Click Edit in Canvas to open the visual editor
3. Drag nodes from the palette and connect them with edges
4. Save and run

Running a Pipeline

Via the Web UI

1. Open a pipeline detail page
2. Optionally set an input path
3. Click Run
4. Monitor progress in real-time via the log viewer

Via the REST API

curl -X POST http://localhost:9433/api/pipelines/my-pipeline/run \
  -H 'Content-Type: application/json' \
  -d '{"inputPath": "/path/to/project"}'

The response returns a jobId for tracking:

{ "jobId": "abc-123-..." }