LangStitch Platform Canvas · SDK · Marketplace · LangTailor IDE

Design AI agent workflows.
Build in Python, Java and Go.

Design on a LangGraph canvas, extend with the Component Designer, share on the marketplace, and ship with the LangTailor IDE (Windows & macOS). Native project generation for Python, Java with Spring AI, and Go, with shared IR and target-specific build and test commands. Two project types: Agentic Development (graphs, skills, deploy) and Plugin Creator (multi-platform packs). Scaffold with langstitch plugin new, build .langstitch-pack.zip with langstitch pack (fail-closed unless --allow-partial). The marketplace also lists personas and prompts alongside connectors, graphs, and MCP packs.

Download LangTailor (Windows & macOS) Explore SDK Marketplace
Python SDK 0.3.2 · PyPI Java compiler 0.2.2 · Maven Central Go · native export Component Designer LangTailor 0.4.0 · Spec 2.2.0 MIT
4 Platform surfaces
82 Marketplace catalog items
Export Python · Java · Go
Release updates

Native runtimes, published packages and clear support boundaries

Explore the September 2026 release and choose a runtime using the tested capabilities.

Python SDK 0.3.2

Real MCP transports, compiler fixes and generated application validation.

Install from PyPI →

Java compiler 0.2.2

Native functions, bounded HTTP tools, authentication and Spring AI runtime fixes.

Find on Maven Central →

LangTailor 0.4.0

Native Python, Java and Go export, language editors, preview and target-specific build/test commands.

Downloads and checksums →

Shared spec 2.2.0

Portable IR with preserved native bodies, target settings and migration fixes.

Package and JSON schemas →

Read the native runtime release notes, feature gap audit and production-readiness evidence.

Questions

Python, Java and Go workflow support

Which languages can I build with LangStitch?

LangTailor 0.4.0 exports Python projects using LangGraph, Java projects using Spring AI, and Go services using the standard library. Python SDK 0.3.2 is on PyPI and Java compiler 0.2.2 is on Maven Central. Go export is included in the IDE; Rust remains planned.

Do all targets support the same features?

Core native functions, HTTP tools, secured invocation and build/test commands are supported. Advanced agents, RAG, MCP, streaming, durable checkpoints, debugging and reverse synchronization differ by target. Python has real MCP clients; Java and Go MCP are not implemented. Read the capability and production-readiness documentation before choosing a target.

Where are releases and verification results documented?

The documentation includes the release notes, capability gaps and production-readiness checks. CI executes generated Python, Java and Go applications on Windows, macOS and Linux. Desktop installers are published for Windows and macOS, with checksums and source provenance. CI does not certify every production deployment or installer interaction.

Ecosystem

Four surfaces, one agent engineering discipline

Design on canvas, extend with SDK, share on marketplace, ship with LangTailor.

01

Visual canvas

LangGraph workflows, asset designers, multi-language export.

Explore features → 02

Multi-language SDK

Python SDK 0.3.2 and Java compiler 0.2.2 are published. LangTailor 0.4.0 includes native Go project export.

sdk.langstitch.com → 03

Marketplace

82 items — 20 connectors · 8 agents · 28 graphs · 16 MCP packs · 10 plugins

Browse catalog → 04

LangTailor IDE

Windows & macOS desktop + VSX — canvas, terminal, git, deploy.

Download →
20Connectors
8Agents
28Graphs
16MCP packs
10Plugins

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

Workflow

From blank canvas to a running agent — in one sitting

Skip weeks of agent boilerplate, RAG wiring, and deploy scripts. Design in LangTailor, export native source in your stack, and validate it with the target toolchain.

01 Design

Map the agent visually

Drag LLM, RAG, Tool, Agent, and SDK components — see the full graph before application code exists.

  • Typed ports catch wiring mistakes early
  • Subgraphs & multi-intent without hand-rolled state
02 Configure

Build assets once, reuse everywhere

Skills, guardrails, personas, and RAG pipelines get dedicated designers — not scattered config files.

  • Vector + hybrid retrieval built in
  • Wire the same skill into any node
03 Ship

Export, run, deploy — no context switch

Export Python, Spring AI Java or Go from the IDE, then build and test with the selected toolchain. Native code bodies and supported runtime settings stay with the graph; unsupported options are explained before export.

  • langsmith.json round-trips back to the IDE
  • Local Build / Run / Deploy toolchain + optional Git sync
1 session canvas → production code
Zero editor extensions to wire up
Full stack graphs · skills · deploy/
MIT download & keep exports
LangTailor downloads & VSX → SDK docs

Your first export is yours forever — no account, no lock-in.

langstitch export Ready to run
Project structure
billing-agent/
├── src/billingagent/       ← Python · shipping today
├── spring/                 ← Spring Boot API gateway (optional)
├── skills/ · guardrails/ · rag/
├── deploy/helm/billing-agent/
├── langstitch.project.json
└── langsmith.json
# Select Python, Java or Go in the platform panel
Terminal · Python runtime
$ cd billing-agent && pip install -r requirements.txt
$ python -m graphs.billing_agent
✓ LangSmith trace started
✓ RAG pipeline loaded · 1,284 chunks
→ Agent ready on :8080
Platform · deploy from IDE
$ docker build -t billing-agent .
$ helm upgrade --install billing ./deploy/helm/billing-agent
✓ Release billing deployed
Features

Build and validate native agent workflows

From canvas design to Docker deploy — design once in IR v2, compile with the SDK, export to the runtime your team uses.

01

Visual canvas

Drag-and-drop LangGraph workflows with typed node ports and live validation.

LLMToolAgentRAGMulti-IntentSubgraph
02

Component Designer

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

03

Asset designers

Skills, guardrails, personas, RAG pipelines — vector / hybrid retrieval.

04

Multi-language export

Python, Java and Go source export with target-specific code editors and build commands. Shared IR preserves native bodies and settings.

05

LangTailor IDE

Canvas + code sync, terminal, git SCM, language tooling, marketplace plugins.

06

Platform ops

Local Build / Run / Test / Deploy toolchain, Git sync, Docker, and Helm — no hosted API required for day-to-day ops.

07

Integrations

MCP Studio, tool registry, remote graphs, A2A, human-in-the-loop, LangSmith, Langfuse.

08

Marketplace

82 items — 20 connectors · 8 agents · 28 graphs · 16 MCP packs · 10 plugins. GitHub sign-in, sync, publish/review at marketplace.langstitch.com.

SDK

Bring LangStitch to your stack

Extend LangStitch with the Component Designer for visual custom nodes and connectors, or use the multi-language SDK to build graphs, skills, guardrails, and policies in code. Python and Java packages are published to PyPI and Maven Central. Native Go export is part of LangTailor; advanced feature coverage differs by target.

Python Available · PyPI

Decorators, typed YAML, GraphBuilder, MCP, guardrails, and the langstitch CLI. Python 3.10+.

Spring AI (Java) Available · Maven Central

IR v2 → Spring Boot + Spring AI via com.langstitch:langstitch-spring-ai:0.2.2. Native Java functions, HTTP tools and secured invocation are available through the compiler and IDE export.

Go Available · LangTailor 0.4.0

Export a standard-library Go HTTP service with native functions, tools and build/test commands. No separate Go SDK package is required.

Rust Planned

Rust is not an available runtime or export target in this release.

Python runtime · available now

langstitch-sdk

pip install langstitch-sdk — decorators, typed application.yaml, CLI scaffold, and LangGraph compilation. Use the shared IR to select Python, Java or Go and consult each target's supported features.

SDK docs PyPI Release notes
Market landscape

How LangStitch compares

LangStitch takes the best ideas from visual agent builders and LangGraph tooling — then centers production support and developer experience. Multi-language project generation — Python SDK, Spring AI Java compiler and native Go export, with buildable source and explicit capability limits.

What most tools optimize for

LangGraph Studio & LangSmith

  • Excellent tracing, evals, and LangGraph-native debugging
  • Cloud-first — assumes LangSmith account and hosted runtime
  • Limited first-class designers for guardrails, business rules, and personas
  • Export is code-oriented; less opinionated multi-module project layout

Langflow & Flowise

  • Fast visual prototyping with broad component libraries
  • Flow JSON is the source of truth — harder to version in Git like app code
  • RAG is often node-level, not a full pipeline designer with hybrid retrieval
  • Production deploy, Helm, and platform API are not the primary UX

Dify, n8n & low-code platforms

  • Strong no-code workflows and connector marketplaces
  • LangGraph semantics and checkpointing are secondary or absent
  • Vendor-hosted runtime — less control over K8s layout and export structure
  • Agent governance (RBAC tools, guardrails as assets) is lighter

Hand-rolled LangGraph repos

  • Full control — but every team reinvents graph layout, RAG wiring, and export scripts
  • Design and code drift apart; onboarding new devs is slow
  • No visual map of skills, guardrails, and personas alongside the graph
  • CI and deploy pipelines are bespoke per project

Side-by-side

Swipe sideways to see the full comparison →

Capability LangStitch LangGraph Studio Langflow / Flowise Dify / n8n
Visual LangGraph canvas ✓ Native ✓ Native Partial / adapter ✗ Own runtime
Downloadable VSX / desktop IDE ✓ Open VSX + LangTailor Cloud account Self-host or cloud Cloud or self-host
Asset designers (skills, guardrails, rules, personas) ✓ First-class sidebar Limited Prompt / component level App / workflow level
RAG pipeline designer (vector / vectorless / hybrid) ✓ Dedicated designer + node Via code / templates Node chains Knowledge base UI
Multi-intent routing node ✓ Built-in classifier node Manual graph edges Router components Branching workflows
Multi-language production export (Git-friendly) ✓ Python · Java · Go native export; shared IR and documented target limits Code export JSON / Python varies API / limited export
Git sync & versioning in IDE ✓ Platform API LangSmith projects ✗ Manual Platform history
Docker build & Helm deploy from UI ✓ Platform panel ✗ ✗ Managed hosting
MCP Studio & tool registry ✓ Built-in MCP via ecosystem Plugins Integrations catalog
Visual custom-component SDK (nodes / connectors / adaptors) ✓ Component Designer + safe codegen Code only Custom components in code Connector SDK
LangSmith / Langfuse observability ✓ Export + hooks ✓ Deep native Plugins Built-in logs
Open source & self-hostable ✓ MIT Partial / cloud ✓ OSS Freemium / OSS tiers
Managed multi-tenant SaaS → Roadmap ✓ LangSmith Cloud ✓ Cloud tiers ✓ Core product
Real-time collaborative editing → Planned ✓ Team features Rare ✓ Shared workspaces
Eval datasets & trace replay (LangSmith depth) → Export hooks today ✓ Best-in-class Basic Moderate
Enterprise SSO / org RBAC → Platform RBAC first ✓ Enterprise Varies ✓ Enterprise tiers
Connector / template marketplace ✓ Built-in marketplace + publish/review LangChain hub ✓ Large libraries ✓ Very large

LangStitch edge — production & DX

  1. Code is the contract — Export Python, Java and Go source, preserve native bodies in shared IR, and review generated projects in PRs. Native reverse code synchronization is not yet supported.
  2. Governance as assets — Skills, guardrails, business rules, and personas are designed once, versioned, and wired on the canvas — not buried in prompt strings.
  3. RAG you can operate — Chunking, embedding, vector / vectorless / hybrid retrieval in a pipeline designer — bound to a RAG Agent node, not a one-off chain.
  4. Platform panel for ops — Git sync, Docker build, Helm deploy, and structured logging hooks — the path from canvas to cluster without a separate toolchain.
  5. Extensible by the community — The SDK Component Designer turns custom nodes, connectors, and adaptors into safe, portable .component.json manifests — shared through a built-in marketplace, not locked to vendor plugins.
  6. Developer-first defaults — Downloadable VSX, LangTailor portable builds, Docker one-liner, Playwright E2E, agent smoke tests, and MIT license — optimized for teams who own their runtime.

Honest gaps — building toward parity + more

  • Managed cloud — No one-click multi-tenant SaaS yet; self-host and Docker first.
  • Deep evals — LangSmith-native dataset runs and trace replay are export hooks today, not full in-IDE eval suites.
  • Live collaboration — Single-editor canvas; no real-time multi-user cursors yet.
  • Enterprise identity — Tool RBAC and guardrails exist; org-wide SSO/SAML is on the roadmap.
  • Marketplace scale — 82 pre-built catalog items (connectors, agents, graphs, MCP packs, personas, prompts, plugins) with publish/review; community growth continues.

Goal: match the table stakes everyone expects, then win on export quality, governance assets, K8s-native deploy, and the Agent Engineering discipline — not on being another locked SaaS canvas.

Summary: LangStitch is for teams shipping LangGraph to production. Python SDK 0.3.2, Java compiler 0.2.2 and LangTailor 0.4.0 native Go export are available. Use LangSmith for traces; use LangStitch to design, export, and deploy the graph your SREs can support.
Contact

Business & training

Platform adoption, team enablement, LangGraph training, and enterprise partnerships — reach out for business and training inquiries.

connect@langstitch.com

Business · training · workshops · partnerships

Privacy: With your consent we store first-party analytics (pages viewed, device/browser, approximate location from IP, referral/UTM) in our database to improve LangStitch. Decline to block tracking cookies. Contact connect@langstitch.com to request deletion.

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