Graph Compose: Hosted Temporal Workflows with Visual Builder and AI

Graph Compose is a hosted platform for orchestrating API workflows on Temporal's durable execution engine. You define workflows as graphs of nodes (HTTP calls, AI agents, iterators, error boundaries) that run as Temporal workflows without managing infrastructure.
Three Ways to Build the Same Graph
Every workflow uses a consistent JSON format that can be created through:
- A React Flow visual builder
- A typed TypeScript SDK (@graph-compose/client)
- An AI assistant that turns plain English into a graph
How It Works
You submit JSON workflows via API to https://api.graphcompose.io/v1/workflows/execute and they run immediately on hosted Temporal. No deploy step, no Temporal cluster to run, no workers to deploy. The platform handles scheduling, retries, error boundaries, and state persistence.
Workflow Example
const workflow = {
"nodes": [
{
"id": "upload_image",
"type": "http",
"dependencies": [],
"http": {
"method": "PUT",
"url": "https://storage.example.com/{{ context.filename }}",
"headers": {
"Content-Type": "{{ context.contentType }}",
"Authorization": "Bearer {{ $secret('storage_token') }}"
},
"body": "{{ context.imageData }}"
},
"activityConfig": {
"retryPolicy": {
"maximumAttempts": 3,
"initialInterval": "1s",
"backoffCoefficient": 2
},
"startToCloseTimeout": "30 seconds"
}
},
{
"id": "generate_thumbnails",
"type": "http",
"dependencies": ["upload_image"],
"http": {
"method": "POST",
"url": "https://images.example.com/process",
"body": {
"source": "{{ results.upload_image.data.url }}",
"sizes": ["400x400", "200x200"]
}
}
}
],
"context": {
"filename": "profile.jpg",
"contentType": "image/jpeg"
}
}
Key Features
- Durable Execution: Workflows survive server restarts, network failures, and process crashes. Temporal replays from the last checkpoint.
- Automatic Retries: Configure retry policies per node with max attempts, backoff intervals, and non-retryable error types.
- Error Boundaries: Wrap nodes in try/catch-style error boundaries with fallback behavior.
- State Persistence: Query workflow state at any point.
- Template Expressions: Nodes reference results from upstream nodes via
{{ results.node_id.data.field }}syntax.
Licensing
Open-core model: the execution foundations and integrations service are AGPL-3.0, while the platform orchestrator, visual builder, and AI assistant are proprietary.
📖 Read the full source: HN AI Agents
👀 See Also

Engram: A Learning Plugin for OpenClaw That Actually Makes Knowledge Stick
Engram is an OpenClaw plugin that uses neuroscience-based techniques like pre-testing, blind grading, and FSRS-4.5 scheduling to ensure you retain what you learn, even as AI agents accelerate your shipping.

Local PII Redaction Skill for OpenClaw Uses GLiNER Model
A new OpenClaw skill intercepts outgoing responses and runs them through the local nvidia/gliner-PII model to detect and redact sensitive information like API keys and PII, replacing them with labels like [API_KEY] and adding removal notices.

Snip tool enables visual communication with AI coding agents
Snip is a free tool that lets developers screenshot, annotate, and draw to show AI agents what they mean visually, while agents can generate diagrams or load images directly through CLI or MCP. Currently runs on Apple Silicon Macs with Mermaid diagram support and HTML support in progress.

100 Popular Apps Reverse-Engineered into Markdown Design Specs for Claude UI Cloning
An open-source repo provides structured markdown design specs for 100 popular iOS apps, optimized for Claude to clone UIs consistently. Key techniques: exact color values, state coverage, spacing scales, and navigation graphs.