Skip to content

Station 0.20.0

Choose a tag to compare

@github-actions github-actions released this 21 Nov 03:00
· 260 commits to main since this release

Station 0.20.0 🚂

Welcome to Station 0.20.0! This release includes improvements to the AI infrastructure platform that makes MCP servers as easy to discover, configure, and deploy as Docker containers.

Installation

One-liner install:

curl -sSL https://getstation.cloudshipai.com | bash

Or with specific version:

curl -sSL https://getstation.cloudshipai.com | bash -s -- --version v0.20.0

Manual download:
Download the appropriate binary for your platform from the assets below.

Go install:

go install github.com/cloudshipai/station/cmd/main@v0.20.0

Changelog

🚀 Features

  • 3b28230: feat(ui): implement paper-matte aesthetic for Agents page with OpenAI-inspired design (@epuerta9)
  • e28b58a: feat(ui): polish agents page - add schedule functionality and clean up UI (@epuerta9)

Other

  • 3711bc1: Add AI-powered agent testing with comprehensive execution tracking (@epuerta9)
  • 5f2acbf: Add Jaeger integration to report generation benchmarks (@epuerta9)
  • 6956218: Add automatic .gitignore generation to stn init (@epuerta9)
  • d92ddce: Add comprehensive LLM-as-judge dataset evaluation system (@epuerta9)
  • 98c263b: Add comprehensive Station MCP Tools reference documentation (@epuerta9)
  • 21c4371: Add comprehensive evaluation and testing guide with LLM-as-judge system (@epuerta9)
  • 67764b6: Add comprehensive multi-agent teams guide with SRE example (@epuerta9)
  • 59282a6: Add quality metrics to PDF reports with LLM-as-judge integration (@epuerta9)
  • 0458990: Add smooth animations and micro-interactions throughout the UI (@epuerta9)
  • 6573612: Add station-demo to gitignore (separate Git repo) (@epuerta9)
  • ad775c7: Apply paper-matte theme to Report Detail header (@epuerta9)
  • bd1ae95: Clean up test environment files (@epuerta9)
  • 01da5fd: Configure OpenCode to use station-demo workspace (@epuerta9)
  • 6e46941: Document preference for opencode-station MCP tools over CLI (@epuerta9)
  • 2cda014: Document successful GitOps workspace proof-of-concept (@epuerta9)
  • c8153b0: Enhance deploy section with MCP endpoint exposure and self-hosted deployment details (@epuerta9)
  • 5acffc8: Fix Agent Performance Criteria undefined error (@epuerta9)
  • e0d3521: Fix Jaeger trace aggregation and add CLI Jaeger integration (@epuerta9)
  • 25106ad: Fix Jaeger trace querying to use correct tag format and Genkit spans (@epuerta9)
  • 1cb039a: Fix Reports page environment filtering and navigation issues (@epuerta9)
  • bc51316: Fix SQLITE_BUSY errors in batch execution by adding shared write mutex to faker sessions (@epuerta9)
  • ba2fbeb: Fix duplicate server keys in MCP Directory (@epuerta9)
  • 2270d36: Fix missing X icon import in Add Server modal (@epuerta9)
  • 4d2ff97: Fix stn init to respect --config flag for workspace initialization (@epuerta9)
  • 9164621: Improve LLM-as-judge evaluation fairness and context-awareness (@epuerta9)
  • a8c65f9: Improve task completion scoring and add queue API variable support (@epuerta9)
  • 4a121c7: Integrate Jaeger trace querying for LLM evaluation tool context (@epuerta9)
  • e900e4d: Merge pull request #79 from cloudshipai/feature/ui-overhaul (@epuerta9)
  • 52bd4fb: Redesign MCP Directory page with paper-matte aesthetic (@epuerta9)
  • 4a0fbcd: Remove duplicate Run Agent button and update graph node styling (@epuerta9)
  • 5617b93: Rewrite README to position Station as AI Agent Orchestration Platform with verified examples (@epuerta9)
  • 0964c8e: Support directory paths in stn init --config flag (@epuerta9)
  • be9fcde: Transform UI from Tokyo Night to OpenAI-inspired light theme across all modals and components (@epuerta9)

What's Station?

Station transforms complex MCP server setup from hours of configuration into a 30-second guided wizard. Think of it as the "package manager for AI agents."

Key Features:

  • 🧙 AI-powered MCP server discovery from GitHub URLs
  • 📋 Interactive configuration wizards with validation
  • 🚀 One-command deployment from repo to running server
  • 🤖 Revolutionary AI agent platform with smart context management
  • 🔒 Enterprise security with local and remote deployment modes

Getting Started

# Initialize Station
stn init

# Discover and deploy any MCP server from GitHub
stn load https://github.com/modelcontextprotocol/servers/tree/main/src/filesystem

# Access admin interface
ssh admin@localhost -p 2222

Full Changelog: v0.19.1...v0.20.0