Visual workflow IDE Canvas · Code · Export · Deploy — one desktop app

Design workflows visually.
Build in Python, Java or Go.

Build agent workflows on a visual canvas, keep their structure in a shared IR document, and export native Python, Java (Spring AI) or Go projects. LangTailor 0.4.0 runs each target's build tools and checks feature compatibility before export. Start with the desktop IDE or the LangTailor Canvas extension.

Download LangTailor (Windows & macOS) Open VSX Release notes
LangTailor v0.4.0 Python · PyPI Spring AI · Maven Central Go · Native export MIT
Download

LangTailor desktop IDE

LangTailor 0.4.0 for Windows x64 and macOS Apple Silicon or Intel. No Linux desktop installer is available in this release.

All releases →
LatestLangTailor v0.4.0Release notesSHA256 checksums

Windows x64 installer

LangTailor-0.4.0-win-x64-setup.exe
v0.4.0
Download

macOS arm64

DMG · Apple Silicon
v0.4.0
Download

macOS x64

DMG · Intel
v0.4.0
Download

Install notes

These installers are unsigned. Windows SmartScreen or macOS Gatekeeper may show a warning or block opening. Check the release source and published checksums before installing. macOS automatic update installation requires a signed application.

  • Windows: Run the setup.exe installer and launch LangTailor from the Start menu.
  • macOS: Open the DMG and drag LangTailor to Applications. If Gatekeeper blocks an unsigned DMG, Right-click → Open, then confirm Open.
  • VS Code / VSCodium: Install from the VS Code Marketplace or Open VSX, or run codium --install-extension langstitch.langtailor-canvas (swap codium for code in VS Code).

Public downloads: github.com/LangStitch/langtailor-releases

Marketplace

Explore the workflow catalog

Live catalog from api.langstitch.com · GitHub sign-in · sync into canvas · publish / review at marketplace.langstitch.com

Open marketplace →
·connectors ·agents ·graphs ·MCP packs ·plugins
Choose your install

Two ways to build with LangStitch

Same canvas, same *.langstitch.json projects — pick the VS Code extension for a focused editor workflow, or LangTailor IDE for advanced control from design through deploy.

LangTailor Canvas · VSX

For VS Code

The full LangStitch graph editor inside VS Code or VSCodium — design workflows and export native projects without switching tools.

Canvas · designers · marketplace · export

  • Visual canvas with LLM, tool, function, router and transformer nodes
  • In-canvas marketplace — browse and sync connectors, graphs, and MCP packs
  • Canvas toolbar with Build / Export and marketplace actions
  • Local project export without a LangStitch cloud account
  • Asset designers — skills, guardrails, personas, RAG pipelines
  • Native Python, Java (Spring AI) and Go export with compatibility checks
  • One-click install from Open VSX — stays in your editor flow
Install from Open VSX Download VSIX v0.4.0 
codium --install-extension langstitch.langtailor-canvas

LangTailor IDE

Full control

The purpose-built agent IDE — everything in VSX, plus local tooling to build, run and test supported workflows.

Canvas · code · terminal · platform · deploy

  • Everything in the VSX canvas extension
  • Canvas and native source previews; Python source synchronization
  • Integrated terminal, git SCM, and language tooling (Python LSP today)
  • Target-specific Build / Run / Test; Python debugging
  • Platform panel — Docker build, Helm deploy, eval runs
  • Marketplace sign-in and plugin sync built in
  • Export Python, Java (Spring AI) and Go projects with their own build commands
Download LangTailor (Windows & macOS) v0.4.0
Start with VSX, grow into IDE anytime. Projects are identical — open the same *.langstitch.json files in LangTailor when you want terminal, git, platform ops, or marketplace without migrating or re-exporting. Download LangTailor →
Language support in 0.4.0

Three native targets, with explicit boundaries

Core workflows use start/end nodes, LLM calls, functions, routers, HTTP tools and response transformers. Each target executes its own generated code.

Java · Spring AI

Export a Maven project with explicit Java implementations, HTTP tools and a Spring Boot service. Use Java 21 and Maven 3.9. The standalone Java compiler 0.2.2 is also on Maven Central.

Native MCP, subgraphs, streaming and checkpoints remain unsupported.

Go · Native service

Export a Go module with explicit Go implementations and a standard-library HTTP service. The generated module requires Go 1.22 or newer; conformance checks run with Go 1.26.

Custom component templates, MCP, subgraphs, streaming and checkpoints remain unsupported.

What the checks cover

Generated core applications are compiled and executed in CI on Windows, macOS and Linux. Desktop installers are offered for Windows x64 and macOS arm64/x64. Build results are evidence for the tested workflows, not a guarantee for arbitrary custom code or a production deployment.

Advanced RAG and agent execution, Java/Go debugging and reverse synchronization still need work. Rust is not an export target in this release. Read the implementation and deployment documentation before choosing a target.

Release history

Versioned releases

Features

LangTailor IDE — beyond the canvas

Design shared IR documents, browse components and export native projects. Run each target's build and test tools locally, with Python MCP and human interrupt support where graph settings allow.

01

Canvas + code

Edit per-language implementations and preview generated source. Python supports source synchronization; Java and Go reverse synchronization remain unsupported.

LLMToolFunctionRouteSubgraph
02

Component Designer

Custom nodes, connectors, adaptors — export as .component.json for marketplace.

03

Asset designers

Author skills, guardrails, personas and RAG assets. Advanced RAG and agent execution are still under development.

04

MCP Studio + integrations

Python exports execute MCP tools. HTTP tools work across Python, Java and Go. Native Java/Go MCP, A2A and remote graph execution remain unsupported.

05

Local Build / Run / Test / Debug / Deploy

Use target-specific build, run and test commands. Python debugging and Docker/Helm controls are available; Java and Go native debugging remains unsupported.

06

Multi-language export

Export Python with SDK 0.3.2, Java with compiler 0.2.2, or a native Go service. Each target validates supported features before generating code.

07

Marketplace

82 items — 20 connectors · 8 agents · 28 graphs · 16 MCP packs · 10 plugins. Sign in once; sync into the IDE.

08

Platform ops

Use terminal and git alongside the designer. Generated deployment files need environment configuration and deployment validation.

Contact

Business & training

Enterprise rollouts, team training, LangGraph workshops, and partnership inquiries — email us and we will follow up.

connect@langstitch.com

Business · training · workshops · partnerships

Privacy: If you accept analytics cookies we record pages viewed, device/browser details, and approximate location (from IP) in our database. Decline to opt out. Email connect@langstitch.com for data requests.

Privacy

Analytics & cookies

How LangStitch uses first-party analytics on this site.

Overview

LangStitch runs first-party analytics on our public sites to understand how visitors use pages and features. We do not use third-party ad trackers or sell visitor data.

What we collect (with consent)

  • Pages viewed, clicks, scroll depth, and time on page
  • Browser, operating system, device type, screen size, and language
  • Approximate location (country/region/city derived from IP)
  • Referrer, UTM campaign parameters, and landing URL
  • A random visitor ID stored in your browser (local storage) and a session ID (session storage)

Your choices

Accept or decline analytics in the cookie banner. If you decline, we do not set tracking cookies or record browsing events. You can clear site data in your browser to reset your choice.

Data requests

Email connect@langstitch.com to ask about your data or request deletion.

Last updated: July 2026