# langchain-ai/open-canvas architecture

> Architecture diagram of the GitHub repository langchain-ai/open-canvas, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2025-07-23). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/langchain-ai/open-canvas
- Repository: https://github.com/langchain-ai/open-canvas

## Overview

##### Project type and purpose
- Open Canvas is a full-stack, monorepo web application for collaborative document editing powered by autonomous agents.
- Frontend: Next.js web app with serverless API routes, React components, Supabase auth.
- Backend/Agents: LangGraph server (“apps/agents”) implements agent graphs (open-canvas, summarizer, reflection, web-search nodes) that orchestrate LLM calls and memory.

##### Top-level components
- Client (Browser)
- Frontend Web App (“apps/web”)
– React UI, Next.js pages & API routes
– Workers for streaming graph results
– Supabase client for auth & user data
- Agent Service (“apps/agents”)
– LangGraph server hosting agent graphs
– Node implementations (generateArtifact, reflection, summarizer, web-search, etc.)
– Shared logic in “packages/shared”
– LLM providers (OpenAI, Anthropic, Google GenAI, Fireworks, Ollama)
– Memory store (built-in LangGraph memory, persisted per thread)
- Shared libraries
– “packages/shared”: prompts, models, types, utils
- Evals service (optional)
– “packages/evals” for testing/benchmarking agents
- External services
– Supabase (authentication, user profiles, maybe DB)
– LLM APIs
– FireCrawl (web scraping), ExaSearch, Groq (transcription)
– LangSmith (observability)

##### Relationships and interactions
A. User → Browser → Next.js frontend
- Renders React UI, handles user input (text/code editor, chats, quick actions).
- Calls Next.js API routes under /api/* for persistence (store/get artifacts, feedback).
- Calls frontend worker (“graph-stream”) to stream agent responses.
B. Frontend API routes → Agent Service (LangGraph)
- The worker or API route sends HTTPS requests to LangGraph server (https://localhost:54367).
- Payload: graph name (e.g. “open-canvas”), initial inputs (current artifact text, user ID, memory tokens).
C. Agent Service internals (deep dive)
1. API receives request → LangGraph CLI routes to appropriate graph.
2. Graph orchestration runs a sequence of nodes:
– generate-artifact: validates input, calls LLM with prompt templates (from packages/shared).
– reflection: optionally runs reflection agent to update memory.
– summarizer, generate-path, thread-title nodes as needed.
3. Each node implementation uses the Langchain SDK (Anthropic, OpenAI clients) or custom tool (FireCrawl).
4. Memory node reads/writes to the graph’s memory store (built-in DB).
5. Final output artifact + updated memory emitted as a streaming response (via SSE).
D. Stream back to Frontend
- Frontend graph-stream worker ingests SSE chunks, updates GraphContext, displays incremental messages in chat UI.
E. Persistence & versioning
- After response complete, Next.js route persists artifact version to Supabase (or LangGraph’s DB).
- Artifacts have version history; UI can fetch previous versions via /api/store/get.
F. Observability & tracing
- LangSmith collects trace data from LangGraph server.

##### Architectural patterns & design principles
- Separation of concerns: UI vs. API vs. agent orchestration.
- Microservice style: Next.js serverless functions vs. standalone LangGraph server.
- Graph-based orchestration of AI workflows (LangGraph).
- Shared code in monorepo packages for types, prompts, models.
- Streaming data flows (SSE/web workers).

##### Extra depth on backend response generation
1. Graph request arrives → LangGraph orchestrator
2. Node A: load state (artifact + memory)
3. Node B: format prompt (packages/shared/prompts)
4. Node C: call LLM API (openai.com, anthropic.com)
5. Node D: parse LLM output → transform to artifact delta
6. Node E: update memory store with reflections
7. Node F: return SSE stream of text tokens back to worker
- Calls to multiple model providers via a “Model Provider” sub-module.
- Plug-in points: adding a new model provider or custom tool node.

## Diagram (Mermaid)

```mermaid
flowchart TD
    %% Client/UI Layer
    subgraph "Client/UI Layer"
        direction TB
        BrowserUI["Browser UI"]:::frontend
        GraphStreamWorker["Graph-Stream Worker\n(streaming SSE)"]:::frontend
    end

    %% Frontend Server/API Layer
    subgraph "Frontend Server/API Layer"
        direction TB
        SupabaseClient["Supabase Client & Middleware"]:::frontend
        NextAPIRoutes["Next.js API Routes"]:::frontend
        ReactComponents["React UI Components"]:::frontend
    end

    %% Agent Service & Data Layer
    subgraph "Agent Service & Data Layer"
        direction TB
        LangGraphServer["LangGraph Server"]:::backend
        GraphDefinition["Open Canvas Agent Graph Definition"]:::backend
        OpenCanvasNodes["Open Canvas Agent Nodes"]:::backend
        ModelProviders["Model Providers\n(OpenAI, Anthropic, etc.)"]:::ai
        MemoryStore["Memory Store"]:::ai
        ReflectionModule["Reflection Agent Sub-module"]:::ai
        SummarizerModule["Summarizer Agent Sub-module"]:::ai
        ThreadTitleModule["Thread-Title Agent Sub-module"]:::ai
        WebSearchModule["Web-Search Agent Sub-module"]:::ai
        LangGraphConfig["LangGraph Configuration"]:::backend
    end

    %% Shared Libraries
    subgraph "Shared Packages"
        direction TB
        SharedPackages["Shared Prompts, Models, Types, Utils"]:::shared
        EvalsService["Evals Service\n(agent testing/benchmarking)"]:::shared
    end

    %% External Services
    subgraph "External Services"
        direction TB
        ExternalSupabase["Supabase"]:::external
        ExternalLLM["LLM APIs"]:::external
        ExternalFireCrawl["FireCrawl, ExaSearch, Groq"]:::external
        ExternalLangSmith["LangSmith (Observability)"]:::external
    end

    %% Connections - User flows
    BrowserUI -->|"user input"| ReactComponents
    ReactComponents -->|"calls API"| NextAPIRoutes
    NextAPIRoutes -->|"uses Supabase"| SupabaseClient
    NextAPIRoutes -->|"HTTP request"| LangGraphServer
    GraphStreamWorker -->|"SSE stream"| LangGraphServer
    LangGraphServer -->|"loads graph"| GraphDefinition
    LangGraphServer -->|"executes nodes"| OpenCanvasNodes
    OpenCanvasNodes -->|"calls provider"| ModelProviders
    OpenCanvasNodes -->|"reads/writes"| MemoryStore
    OpenCanvasNodes -->|"invokes"| ReflectionModule
    OpenCanvasNodes -->|"invokes"| SummarizerModule
    OpenCanvasNodes -->|"invokes"| ThreadTitleModule
    OpenCanvasNodes -->|"invokes"| WebSearchModule
    LangGraphServer -->|"config via"| LangGraphConfig
    SharedPackages --> ReactComponents
    SharedPackages --> OpenCanvasNodes
    EvalsService --> OpenCanvasNodes
    SupabaseClient --> ExternalSupabase
    ModelProviders --> ExternalLLM
    WebSearchModule --> ExternalFireCrawl
    LangGraphServer --> ExternalLangSmith

    %% Click Events
    click BrowserUI "https://github.com/langchain-ai/open-canvas/tree/main/apps/web"
    click ReactComponents "https://github.com/langchain-ai/open-canvas/tree/main/apps/web/src/components"
    click NextAPIRoutes "https://github.com/langchain-ai/open-canvas/tree/main/apps/web/src/app/api"
    click GraphStreamWorker "https://github.com/langchain-ai/open-canvas/tree/main/apps/web/src/workers/graph-stream"
    click SupabaseClient "https://github.com/langchain-ai/open-canvas/tree/main/apps/web/src/lib/supabase"
    click LangGraphServer "https://github.com/langchain-ai/open-canvas/tree/main/apps/agents"
    click GraphDefinition "https://github.com/langchain-ai/open-canvas/blob/main/apps/agents/src/open-canvas/index.ts"
    click OpenCanvasNodes "https://github.com/langchain-ai/open-canvas/tree/main/apps/agents/src/open-canvas/nodes"
    click ReflectionModule "https://github.com/langchain-ai/open-canvas/tree/main/apps/agents/src/reflection"
    click SummarizerModule "https://github.com/langchain-ai/open-canvas/tree/main/apps/agents/src/summarizer"
    click ThreadTitleModule "https://github.com/langchain-ai/open-canvas/tree/main/apps/agents/src/thread-title"
    click WebSearchModule "https://github.com/langchain-ai/open-canvas/tree/main/apps/agents/src/web-search"
    click SharedPackages "https://github.com/langchain-ai/open-canvas/tree/main/packages/shared"
    click EvalsService "https://github.com/langchain-ai/open-canvas/tree/main/packages/evals"
    click LangGraphConfig "https://github.com/langchain-ai/open-canvas/blob/main/langgraph.json"

    %% Styles
    classDef frontend fill:#E0F7FA,stroke:#0288D1,color:#000;
    classDef backend fill:#E8F5E9,stroke:#388E3C,color:#000;
    classDef ai fill:#FFF3E0,stroke:#F57C00,color:#000;
    classDef shared fill:#EDE7F6,stroke:#6A1B9A,color:#000;
    classDef external fill:#F5F5F5,stroke:#9E9E9E,color:#000;
```
