steward-cli 0.21.0 → 0.23.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (4) hide show
  1. package/README.md +34 -93
  2. package/dist/cli.js +12983 -70751
  3. package/package.json +2 -10
  4. package/dist/steward +0 -4
package/README.md CHANGED
@@ -1,36 +1,16 @@
1
1
  # Steward
2
2
 
3
- [![Version](https://img.shields.io/badge/version-v0.21.0-D77757.svg)](https://github.com/sapirrior/steward/releases)
4
- [![npm](https://img.shields.io/npm/v/steward-cli.svg?color=373737)](https://www.npmjs.com/package/steward-cli)
5
- [![License](https://img.shields.io/badge/license-MIT-blue.svg)](LICENSE)
6
- [![Runtime](https://img.shields.io/badge/runtime-Bun-fbf0df.svg?logo=bun)](https://bun.sh)
7
- [![Platforms](https://img.shields.io/badge/platform-Linux%20%7C%20macOS%20%7C%20Termux%20%7C%20Windows-lightgrey.svg)](#get-started)
3
+ [![npm](https://img.shields.io/npm/v/steward-cli?color=E5533D&label=steward&style=flat-square)](https://www.npmjs.com/package/steward-cli)
4
+ [![license](https://img.shields.io/badge/license-MIT-7A88CF?style=flat-square)](LICENSE)
5
+ [![chat](https://img.shields.io/badge/terminal-AI%20companion-30B08F?style=flat-square)](#get-started)
8
6
 
9
- Steward is an interactive AI engineering assistant for the terminal. Built with a lightweight native runtime, Steward pairs with you to explore codebases, refactor architectures, run commands, and implement features with full human-in-the-loop control across local and frontier models.
7
+ Steward is an interactive AI engineering assistant that lives in your terminal, understands your codebase, and pairs with you to build, refactor, and debug software through natural language.
10
8
 
11
9
  ---
12
10
 
13
11
  ## Get Started
14
12
 
15
- ### Installation
16
-
17
- Install Steward globally using npm or Bun:
18
-
19
- ```bash
20
- # npm (Recommended)
21
- npm install -g steward-cli
22
-
23
- # Bun
24
- bun add -g steward-cli
25
- ```
26
-
27
- Or execute directly with npx:
28
-
29
- ```bash
30
- npx steward-cli
31
- ```
32
-
33
- #### Standalone Scripts
13
+ ### Quick Install
34
14
 
35
15
  **macOS, Linux & Termux:**
36
16
  ```bash
@@ -42,91 +22,52 @@ curl -fsSL https://raw.githubusercontent.com/sapirrior/steward/main/installer/in
42
22
  irm https://raw.githubusercontent.com/sapirrior/steward/main/installer/install.ps1 | iex
43
23
  ```
44
24
 
45
- **Build from Source:**
25
+ **Package Managers:**
46
26
  ```bash
47
- git clone https://github.com/sapirrior/steward.git
48
- cd steward
49
- bun install
50
- bun run start
51
- ```
52
-
53
- ---
54
-
55
- ## Core Capabilities
56
-
57
- - **Human-in-the-Loop Engineering:** Steward keeps you in control with explicit permission gates for filesystem mutations and bash commands.
58
- - **First-Class Local Model Support:** Optimized for local model execution (Ollama, LM Studio, vLLM, Qwen, DeepSeek) as well as cloud frontier APIs.
59
- - **Lightweight Native Architecture:** Handcrafted zero-overhead runtime with sub-millisecond startup and minimal system resource usage.
60
- - **Synchronized Terminal UI:** Built with Mode 2026 synchronized output, flicker-free differential rendering, and intra-line word diffs.
61
- - **Lossless Rewind Checkpoints:** Every file mutation is automatically tracked with Content-Addressed Storage (CAS) hashes, allowing one-click turn rollbacks.
62
-
63
- ---
64
-
65
- ## Privacy & Data Security
66
-
67
- Steward operates on a strictly local-first architecture:
68
-
69
- - **Zero Telemetry:** Steward collects no analytics, tracking data, or diagnostic logs.
70
- - **Local Persistence:** All session history, configuration, and state are stored locally on your machine.
71
- - **Direct Inference:** Network requests are made directly to your configured provider or local server endpoint via your own API credentials, governed entirely by your provider's terms.
72
-
73
- ---
27
+ # npm
28
+ npm install -g steward-cli
74
29
 
75
- ## Provider Setup & Configuration
30
+ # Bun
31
+ bun add -g steward-cli
32
+ ```
76
33
 
77
- Steward supports local endpoints and major cloud providers. Configure your environment using environment variables or a `.env` file:
34
+ ### Usage
78
35
 
79
- ### Local Models (Ollama, LM Studio, vLLM)
36
+ Navigate to any project directory and run:
80
37
 
81
38
  ```bash
82
- export CUSTOM_API_URL="http://localhost:11434/v1"
83
- export CUSTOM_API_MODEL_NAME="qwen2.5-coder:32b"
84
- export CUSTOM_API_KEY="ollama"
39
+ steward
85
40
  ```
86
41
 
87
- ### Cloud Providers
42
+ Authenticate with your provider of choice using `/login` (Anthropic, OpenRouter, GitHub Copilot) or standard API keys.
88
43
 
89
- ```bash
90
- # Anthropic
91
- export ANTHROPIC_API_KEY="sk-ant-..."
92
-
93
- # Google Gemini
94
- export GEMINI_API_KEY="AIzaSy..."
95
-
96
- # OpenAI
97
- export OPENAI_API_KEY="sk-..."
44
+ ---
98
45
 
99
- # DeepSeek
100
- export DEEPSEEK_API_KEY="sk-..."
46
+ ## Supported Providers & Models
101
47
 
102
- # xAI
103
- export XAI_API_KEY="xai-..."
48
+ Steward natively connects to frontier APIs, gateways, and local offline inference servers:
104
49
 
105
- # Mistral AI
106
- export MISTRAL_API_KEY="your-mistral-api-key"
107
-
108
- # OpenRouter
109
- export OPENROUTER_API_KEY="sk-or-v1-..."
110
- ```
50
+ | Provider | Authentication | Default Model | Key / Config |
51
+ | :--- | :--- | :--- | :--- |
52
+ | **Anthropic** | Browser OAuth / API Key | `claude-3-7-sonnet-20250219` | `/login anthropic` or `ANTHROPIC_API_KEY` |
53
+ | **OpenAI** | API Key | `gpt-4o-mini` | `OPENAI_API_KEY` |
54
+ | **Google Gemini** | API Key | `gemini-2.5-flash` | `GEMINI_API_KEY` |
55
+ | **OpenRouter** | Browser OAuth / API Key | `meta-llama/llama-3.3-70b-instruct:free` | `/login openrouter` or `OPENROUTER_API_KEY` |
56
+ | **GitHub Copilot** | Device Code OAuth | `gpt-4o` | `/login github-copilot` |
57
+ | **Ollama (Local)** | None | `qwen2.5-coder:latest` | `OLLAMA_BASE_URL` (default `http://localhost:11434/v1`) |
58
+ | **DeepSeek** | API Key | `deepseek-chat` | `DEEPSEEK_API_KEY` |
59
+ | **Groq** | API Key | `llama-3.3-70b-versatile` | `GROQ_API_KEY` |
60
+ | **xAI / Grok** | API Key | `grok-2-1212` | `XAI_API_KEY` |
61
+ | **Mistral AI** | API Key | `codestral-latest` | `MISTRAL_API_KEY` |
62
+ | **Custom** | Optional API Key | `default` | `CUSTOM_API_URL` & `CUSTOM_API_KEY` |
111
63
 
112
64
  ---
113
65
 
114
- ## Slash Commands
115
-
116
- | Command | Description |
117
- | :--- | :--- |
118
- | `/rewind` | Revert workspace modifications and conversation state to any prior turn |
119
- | `/model` | Switch active model or provider endpoint |
120
- | `/mode` | Cycle through or set active chat mode (`normal`, `chat`, `review`, `build`) |
121
- | `/theme` | Change terminal color theme (`dark`, `light`, `dracula`, `dark-ansi`, `light-ansi`) |
122
- | `/effort` | Adjust model reasoning effort level (`none`, `low`, `medium`, `high`, `xhigh`) |
123
- | `/sessions` | Browse, resume, or delete saved session transcripts |
124
- | `/clear` | Clear the current conversation and start a new session |
125
- | `/help` | View help manual and keyboard shortcuts |
126
- | `/exit` | Exit Steward (`Ctrl+C Ctrl+C`) |
66
+ ## Contributing
127
67
 
128
- ---
68
+ We welcome contributions! Please see [CONTRIBUTING.md](CONTRIBUTING.md) for development setup and guidelines.
129
69
 
130
70
  ## License
131
71
 
132
72
  MIT License © 2026 [sapirrior](https://github.com/sapirrior)
73
+