open-gitagent

open-gitagent / langship.sh

Public

The platform for shipping, governing AI agents — framework-agnostic, multi-runtime, GitOps-native, self-hosted.

www.langship.sh
14
1
100% credibility
Found May 14, 2026 at 14 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Go
AI Summary

Langship is a self-hosted web app and command-line tool for defining visual pipelines to build, scan, approve, and deploy AI agent applications to various runtimes like AWS Bedrock or Kubernetes.

How It Works

1
📰 Discover Langship

You hear about Langship, a friendly tool that helps everyday people safely build, check, and launch their AI helpers without getting stuck on tech details.

2
💻 Start everything up

With one easy command, you get the web dashboard and all the helper pieces running on your computer, ready to use right away.

3
🔗 Connect your project

Add the link to your AI helper's folder online, and give it a special password if needed so it stays private.

4
🎨 Design your safety flow

Drag simple blocks like 'build it', 'check for issues', 'get approval', and 'launch' to create a step-by-step plan for your helper.

5
🏷️ Set up stages

Name your stages like 'testing' or 'live' and link your flow to them so promotions happen automatically.

6
▶️ Run and watch live

Push the start button or connect a code change trigger, then watch the colorful dashboard show each step happening in real time.

🎉 Your helper is live!

Your AI agent deploys smoothly across places like cloud services, with built-in checks keeping everything safe and auditable.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 14 to 14 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is langship.sh?

Langship.sh is a self-hosted GitHub agent platform for shipping and governing AI agents across any framework like LangGraph, LangChain, or CrewAI. You build pipelines as drag-and-drop graphs—trigger on push, build OCI containers, scan for secrets, gate with approvals, promote via GitOps PRs, and deploy to AWS Bedrock, Kubernetes, or Vertex—all from one YAML definition stored in git. The Go backend pairs with a Next.js UI for live canvas views and a Python CLI for terminal ops, spinning up via Docker Compose with Mongo, Restate, and BuildKit.

Why is it gaining traction?

It stands out as framework-agnostic with GitOps-native promotion: a single pipeline ships agents to multiple runtimes without vendor lock-in, and durable Restate journaling survives crashes for reliable approvals. Developers hook on the 5-minute quickstart to a "green run," CLI commands like `langship agents trigger` plus SSE live logs, beating fragmented agent deploys. Self-hosted keeps creds and history air-gapped.

Who should use this?

AI agent builders shipping prototypes to prod, like teams iterating LangGraph graphs who need scans, approvals, and Bedrock deploys without ArgoCD sprawl. Platform engineers at startups wanting a GitHub platform sh for agent CI/CD, or ops folks enforcing policy gates on multi-runtime agent fleets.

Verdict

Try it if you're early on agent infra—solid README quickstart and Apache 2.0 make it low-risk, despite 14 stars and 1.0% credibility signaling alpha maturity. Polish the roadmap for stubs like K8s deploys, but it's a fresh framework-agnostic win for self-hosted agent shipping.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.