Workflow Blueprints

Visual Automation That Thinks

Design multi-step automation blueprints on a drag-and-drop canvas. Unlike static BPM tools, every workflow can embed LLM reasoning, agent orchestration, and knowledge retrieval — turning fixed pipelines into intelligent processes.

Low-Code Logic

Visual Workflow Editor

A drag-and-drop canvas built on React Flow v12. Design multi-step automation blueprints visually — connect nodes, define conditions, and wire up data transformations without writing orchestration code.

Drag-and-drop canvas with zoom, pan, and minimap navigation
25 built-in node types across four categories
Inline validation badges show configuration errors before execution
Real-time SSE status streaming during execution
INPUT_TRIGGERLLM_Analysisreason · contextSQL_Queryfetch · rows

25 Node Types, Four Categories

Each node type is purpose-built for a specific automation task. Combine them freely to model anything from a simple LLM chain to a complex multi-system orchestration.

Core

Foundational building blocks for every workflow

Start / End
LLM
ConditionBranch
+1 more...

Integration

Connect workflows to external systems, models, and tools

Agent
KnowledgeRetrieval
Connector
+3 more...

Data

Transform, extract, and aggregate data between nodes

VariableAssign
TemplateTransform
CodeExecution
+2 more...

Flow Control

Manage execution flow, loops, and sub-processes

Iterator
Loop
ListOperation
+3 more...
Live Execution #1842
SUCCESS
Input_Parse12ms
Knowledge_RAG450ms
LLM_Reasoning1.2s
Notify_Result85ms

Topological Execution Engine

Workflows execute in dependency order determined by topological sort. Independent branches run concurrently. Condition nodes dynamically select execution paths at runtime.

Dependency-Ordered Execution

Nodes execute only when all upstream dependencies complete. Independent paths run in parallel automatically.

Cross-Node Variable Passing

Every node output is addressable by downstream nodes via structured variable references — no manual data wiring needed.

Real-Time SSE Streaming

Execution status, node outputs, and errors stream to the UI in real time via Server-Sent Events.

Error Strategies

Per-node error handling: STOP_WORKFLOW halts everything, CONTINUE skips the failed node, FAIL_BRANCH stops only the affected branch.

Three Trigger Modes

Workflows can be triggered manually, on a schedule, or via API — each with independent configuration.

Manual

Run workflows directly from the UI with custom input parameters. Ideal for ad-hoc tasks and testing.

Scheduled (Cron)

Configure per-workflow cron schedules with timezone support and default input values. Workflows run automatically at specified intervals.

API Trigger

Each workflow gets a dedicated API key (wf_ prefix) for external execution. No user authentication required — integrate with CI/CD, webhooks, or external systems.

Batch Execution

Submit up to 100 input sets in a single POST /batch-run request. Configure parallelism from 1 to 10 concurrent executions. Ideal for bulk processing — document classification, data enrichment, report generation.

Max Inputs per Batch100
Parallelism Range1 - 10
API EndpointPOST /batch-run
cURL Example
curl -X POST /batch-run \
  -H "Authorization: Bearer wf_..." \
  -d '[{ "q": "..." }, ...]'

Execution Trace Viewer

Debug workflow runs with a timeline-based trace viewer. Step through each node's execution, inspect input/output variable snapshots, and identify exactly where issues occurred.

Timeline view showing node execution order and duration
Per-node variable snapshots — inspect exact inputs and outputs at each step
Error details with stack traces for failed nodes
Execution log viewer with live SSE event stream
v1.2.0
v1.2.1 (Current)

Version Management & Diff

Side-by-side visual diff with color-coded change indicators
Node additions, deletions, and modifications highlighted automatically
Edge connection changes tracked and displayed

Developers

Explore our Source Available code on GitHub, contribute to the connector ecosystem, or integrate FIM One into your own applications.

git clone https://github.com/fim-ai/fim-one.git && ./start.sh

Enterprise

Need private deployment, custom connectors, or professional support? Our team is ready to help you scale your AI transformation.

Private Deploy & Isolation
SSO & Audit Logs
1-on-1 Dedicated Support
SLA Availability Guarantee