MORNINGSTAR

morningstar
╔══════════════════════════════════════════════════════════╗
M O R N I N G S T A R
Terminal AI Coding Assistant
╠──────────────────────────────────────────────────────────╣
Model morningstar-14b [local]
Project my-app (TypeScript / Next.js)
Branch main
Perms ask
╠──────────────────────────────────────────────────────────╣
Tools read · write · edit · bash · grep · glob
ls · git · web · fetch · gh
Agents code · debug · review · refactor · architect
╚══════════════════════════════════════════════════════════╝
/help · /features · /model · /agents · /imagine · /skill:list
> Analysiere dieses Projekt und finde Bugs
Live Demo

Watch it think.

morningstar — ~/my-app
Capabilities

Everything you need. Nothing you don't.

36 features. 28 brand-new capabilities rivaling Claude Code.

7 AI Providers

DeepSeek, OpenAI, Anthropic, Google, Ollama, Groq, OpenRouter — switch mid-session with /model

12 Built-in Tools

read, write, edit, delete, bash, grep, glob, ls, git, web, fetch, gh — with undo, diffs & permissions

NEW

Skills System

Reusable prompt-based skills loaded from .md files — auto-trigger on keywords, stack with agents

NEW

MCP Integration

Model Context Protocol — connect external tool servers, expose Morningstar as MCP server

NEW

8 Permission Modes

auto, ask, strict, bypassPermissions, acceptEdits, plan, dontAsk, delegate — granular access control

NEW

Extended Thinking

4 effort levels (low → ultra) with provider-specific thinking params — Anthropic, OpenAI, DeepSeek, Google

NEW

File-Based Agents

Define agents as .md files with YAML frontmatter — custom tools, models, temperature per agent

NEW

Rules System

Project rules from .md files with @import, path patterns, and priority — auto-injected into system prompt

NEW

Enhanced Hooks

14 hook events — preToolExecution, postFileWrite, preBash, sessionStart & more with JSON decision control

NEW

Agent Teams

Orchestrate multiple agents with roles (lead, worker, reviewer) — shared task queue & sequential execution

NEW

Native Function Calling

API-level tool definitions — OpenAI tools format, Anthropic tool_use blocks, Google functionCall — no regex parsing

NEW

Real Token Counting

Actual token usage from provider APIs with reasoning tokens, cache hits — not estimates. Per-model cost breakdown

NEW

Task Progress Checklist

Live Claude Code-style checklist — ✔ completed, ◼ in-progress, ◻ pending — with duration, tokens, turn counter

NEW

Auto-Test After Changes

Detects vitest, jest, pytest, cargo test, go test — runs relevant tests after file writes and feeds results back to AI

NEW

Sub-Agent Delegation

/delegate <agent> <task> — specialized agents execute autonomously with own conversation, tools, and turn limits

NEW

Parallel Tool Execution

Independent tool calls execute concurrently — up to 10 agentic turns with auto-fix on errors

Context Radar

Real-time context window monitor — see token usage per segment (System, Chat, Code, Tools) with color-coded bar

Animated Diff

Code morphing animation — old lines fade out in red, new lines type in green with typewriter effect

Dependency Graph

ASCII architecture visualization with /graph — scans imports, builds layered topology with box-drawing

NEW

Sandbox Mode

OS-level sandboxing — macOS sandbox-exec, Linux firejail — restrict filesystem access to project directory

NEW

Plugin System

Discover, install & manage plugins — registerTool, registerAgent, registerSkill, registerHook API

VS Code Highlighting

Shiki-powered syntax highlighting with the vitesse-dark theme — same engine as VS Code

Memory & History

Persistent notes + input history across sessions, todo system with priorities, save & restore conversations

NEW

Smart Auto-Fallback

Cloud API blocked? Auto-fallback to local models. Local too weak? Upgrades to cloud. Seamless multi-tier routing

NEW

Git Checkpoints

Create, list & restore checkpoints — pairs git state with conversation history for full rewind

NEW

AI Image Generation

Nano Banana Pro (Gemini 3) + RealVisXL local fallback — 6 models, auto-fallback, no content filters, saves to ~/Downloads

NEW

File Watcher

Watches project files for external changes — auto-triggers tests, notifies AI of modified files with debouncing

NEW

Conversation Branching

Fork conversations at any point, create named branches, switch between them, merge insights back

NEW

Streaming Diffs

Real-time diff preview as the AI generates edit operations — see changes character by character

NEW

Self-Update

/update check + /update run — pulls latest from GitHub, rebuilds, notifies of new versions

NEW

Web Dashboard

Browser-based dashboard at /dashboard — session stats, cost tracking, tool usage, conversation history

NEW

Multi-File Context

10 @-mention types with smart chunking — @files:glob, @diff:file, @tree — context budget management with auto-truncation

NEW

Smart PR Review

/pr-review — fetch, analyze, and review GitHub Pull Requests via gh CLI with structured severity ratings

NEW

Prompt Caching

Provider-aware prompt caching — Anthropic cache_control breakpoints, OpenAI prefix caching, Google context caching with stats

NEW

Terminal Multiplexer

Split-pane layouts (default, split, triple, quad) with focus navigation, scrolling, and Unicode box-drawing borders

7 Themes

Default, Ocean, Hacker, Sunset, Nord, Rose, Claude — from bold colors to minimalist monochrome

Providers

7 Providers. Any Model.

DeepSeek
R1 Reasoner, Chat
--model deepseek-reasoner
OpenAI
o3, o4-mini, GPT-4.1, Codex
--model o3
Anthropic
Claude Opus 4, Sonnet 4
--model claude-opus-4-20250514
Google
Gemini 2.5 Flash, 3 Pro + Nano Banana Image
--model gemini-2.0-flash
Ollama
Morningstar 13B, LLaMA 3, Qwen, any local
--model morningstar
Groq
LLaMA 3.3, Mixtral
--model llama-3.3-70b-versatile
OpenRouter
Any model via API
--model openrouter/auto
$ morningstar --model o3
Provider auto-detected: OpenAI
Open Source Models

Our Models on Hugging Face

3 purpose-built open-weight models for code generation, architecture design, and vision. Free to download, fine-tune, and deploy. Run locally with Ollama.

huggingface.co/kurdman991

Benchmark Comparison

Morningstar 14B vs popular open-source coding models

Morningstar 14BQwen2.5 14BCodeLlama 34BDeepSeek 33B
HumanEval
82.3
79.9
53.7
56.1
HumanEval+
76.8
74.2
47.0
49.4
MBPP
76.5
74.1
56.2
60.8
MBPP+
65.2
62.8
47.1
51.3
MultiPL-E
68.2
65.3
38.4
42.1
DS-1000
47.8
45.1
32.5
37.2

Morningstar 14B beats the base Qwen2.5-14B across all benchmarks, and outperforms CodeLlama-34B and DeepSeek-33B by a wide margin despite being smaller.

Built for Developers

100+ Languages
C, Rust, Go, Python, TS, Solidity & more
128K Context
Read entire codebases in one prompt
Uncensored
No safety filters or content refusal
Apache 2.0
Free for commercial & personal use
Languages

40+ Programming Languages

First-class support for every major language. Morningstar understands idioms, conventions, and best practices for each one.

Python
TypeScript
JavaScript
Node.js
Deno
Bun
Go
Rust
C
C++
C#
Java
Kotlin
Swift
Objective-C
PHP
Ruby
Scala
Dart
Elixir
Haskell
Lua
Groovy
Shell
PowerShell
R
MATLAB
SQL
GraphQL
Solidity
Vyper
Assembly
WebAssembly
GDScript
Nim
Crystal
Zig
COBOL
Fortran
Hack
Internationalization

19 Interface Languages

The CLI speaks your language. Full UI translation including prompts, errors, and help text. RTL support included.

🇬🇧
English
English
🇺🇸
US English
English
🇩🇪
German
Deutsch
🇫🇷
French
Français
🇪🇸
Spanish
Español
🇮🇹
Italian
Italiano
🇸🇦
Arabic
العربية
RTL
🟡
Kurdish
کوردی
RTL
🇹🇷
Turkish
Türkçe
🇵🇹
Portuguese
Português
🇳🇱
Dutch
Nederlands
🇷🇺
Russian
Русский
🇨🇳
Chinese
中文
🇯🇵
Japanese
日本語
🇰🇷
Korean
한국어
🇨🇭
Swiss German
Schwiizerdütsch
🇦🇹
Austrian
Österreichisch
🇹🇼
Chinese (Trad.)
繁體中文
🇬🇧
UK English
British
CLI Design

Built for Speed

A CLI experience designed for power users. Every interaction is optimized for developer velocity.

Interactive Mode

Full autocomplete, syntax highlighting, and inline suggestions.

Tab completion for commands, flags, and file paths

Streaming Output

Watch AI responses stream in real-time with live formatting.

Token-by-token rendering with markdown support

Live Diff View

Review every change before it is applied with highlighted diffs.

Claude Code-style inline diffs with line numbers

Keyboard Shortcuts

Power-user shortcuts for maximum velocity.

ctrl+o expand · ctrl+c cancel · Esc+Esc rewind · Ctrl+R search

Extended Thinking

4 effort levels with provider-specific thinking parameters.

/effort low|medium|high|ultra · /ultrathink

Checkpoints & Rewind

Git-based checkpoints paired with conversation state.

/checkpoint create · /rewind N · Esc+Esc undo

Vim Mode

Native Vim keybindings for the input editor.

Normal, Insert, motions, and ex commands

! Bash Mode

Prefix with ! to execute shell commands directly.

!npm test · !git status · instant execution

Skills & Rules

Reusable .md-based skills and project rules with auto-trigger.

/skill:list · /rules list · YAML frontmatter

MCP & Plugins

Connect external tool servers and install community plugins.

/mcp add · /plugins install · Plugin API

Sandbox Mode

OS-level sandboxing restricts filesystem access.

--sandbox · macOS sandbox-exec · Linux firejail

Export & Copy

Export conversations as markdown, copy last response.

/export · /copy · /rename session

AI Image Generation

Nano Banana Pro (Gemini 3) + local RealVisXL with auto-fallback.

/imagine <prompt> · 6 models · no content filters

Persistent History

Input history persists across sessions. Arrow keys to navigate.

Ctrl+R search · case-insensitive cd · ~/Downloads

Auto-Test

Detects test runners and runs relevant tests after file changes.

vitest · jest · pytest · cargo test · go test

Sub-Agent Delegation

Delegate tasks to specialized agents with own conversation context.

/delegate <agent> <task> · /delegate:list

Task Progress

Live checklist tracking agentic loop turns, tokens, and duration.

✔ done · ◼ running · ◻ pending · up to 10 turns

Native Function Calling

API-level tool definitions instead of regex parsing.

OpenAI tools · Anthropic tool_use · Google functionCall

File Watcher

Auto-detects directories, watches for external changes.

/watch · debounced · auto-test trigger

Conversation Branching

Fork, switch, and merge conversation branches.

/branch · /branch switch · /branch merge

Self-Update

Check for updates and pull latest from GitHub.

/update check · /update run · auto-rebuild

Web Dashboard

Browser-based session dashboard with live stats.

/dashboard · costs · tools · conversation

Multi-File Context

10 @-mention types with smart chunking and context budget.

@files:glob · @diff:file · @tree · 32k budget

Smart PR Review

Fetch, analyze, and review GitHub PRs with severity ratings.

/pr-review <number> · gh CLI · auto-post comments

Prompt Caching

Provider-aware caching with hit stats and cost savings.

/cache · Anthropic · OpenAI · Google

Terminal Multiplexer

Split-pane layouts with focus navigation and scrolling.

/split · default · split · triple · quad
Security

Security First

Your code is your most valuable asset. Morningstar treats it that way.

8 Permission Modes

From strict approval to full bypass — auto, ask, strict, bypassPermissions, acceptEdits, plan, dontAsk, delegate.

OS-Level Sandbox

macOS sandbox-exec and Linux firejail restrict filesystem access to your project directory only.

Encrypted Session Memory

Session data is encrypted at rest. Conversation history never leaves your machine.

Local-Only Mode

Run entirely offline with Ollama. Zero data transmitted to external servers.

Injection Prevention

Built-in prompt injection detection and sanitization for all AI interactions.

Hook-Based Access Control

14 hook events with JSON decision control — pre/post validation on every tool execution.

Per-Project Settings

Isolated allow/deny lists, tool restrictions, and command filters per project via .morningstar/settings.local.json.

Install

Get started in 30 seconds.

Requires Node.js 22+

terminal
$ git clone https://github.com/morningstarnasser/morningstar-cli.git
$ cd morningstar-cli
$ npm install && npm run build && npm link
$ morningstar
Welcome to Morningstar AI
Environment
export DEEPSEEK_API_KEY=sk-...
.env File
OPENAI_API_KEY=sk-proj-...
Interactive
First launch prompts you