@emend-ai/utim 1.46.22 → 1.46.28
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.
- package/README.md +124 -82
- package/package.json +1 -1
package/README.md
CHANGED
|
@@ -1,82 +1,124 @@
|
|
|
1
|
-
# UTIM CLI Agent: Enterprise Coder Assistant
|
|
2
|
-
|
|
3
|
-
UTIM is an agentic developer CLI assistant designed to automate coding tasks directly inside your local terminal, featuring robust safety controls, self-healing quality gates, and local-first semantic memory.
|
|
4
|
-
|
|
5
|
-
---
|
|
6
|
-
|
|
7
|
-
## 🚀 Quick Start
|
|
8
|
-
|
|
9
|
-
### 1. Installation
|
|
10
|
-
Install the package from your local source directory:
|
|
11
|
-
|
|
12
|
-
```bash
|
|
13
|
-
# Basic installation
|
|
14
|
-
pip install .
|
|
15
|
-
|
|
16
|
-
# Recommended: Full installation (includes semantic vector RAG & web search)
|
|
17
|
-
pip install ".[full]"
|
|
18
|
-
```
|
|
19
|
-
|
|
20
|
-
### 2. Provider & Model Configuration
|
|
21
|
-
Start the chat session:
|
|
22
|
-
```bash
|
|
23
|
-
utim
|
|
24
|
-
```
|
|
25
|
-
On first run, UTIM checks for `.utim/config.json`. If it does not exist, it runs a provider and model configuration wizard. You can configure:
|
|
26
|
-
- **Default Providers**: OpenRouter, OpenAI, custom servers, etc.
|
|
27
|
-
- **Model Picker**: Press `Ctrl+M` in the chat terminal at any time to configure, pick, add, or delete LLMs.
|
|
28
|
-
|
|
29
|
-
---
|
|
30
|
-
|
|
31
|
-
## ⚡ Main CLI Commands
|
|
32
|
-
|
|
33
|
-
- **`utim`**: Starts the interactive chat terminal (TUI).
|
|
34
|
-
- **`utim task "<prompt>"`**: Executes a single task autonomously from the command line and exits. In an interactive terminal (`stdin` is a TTY) file writes and commands prompt for confirmation; when piped/scripted they run in auto-accept mode.
|
|
35
|
-
- **`utim --dry-run`**: Starts the session in **Dry-Run Mode** (all code modifications and shell commands are simulated, not written/executed).
|
|
36
|
-
- **`utim --sandbox`**: Runs all mutating shell command proposals in the intelligent local sandbox (untrusted commands will block until approved).
|
|
37
|
-
- **`utim doctor`** / **`utim init`** / **`utim reset`**: Administrative commands for state diagnosis, initialization, and factory resets.
|
|
38
|
-
|
|
39
|
-
---
|
|
40
|
-
|
|
41
|
-
## 🛠️ In-Chat Slash Commands
|
|
42
|
-
|
|
43
|
-
Inside the interactive chat terminal, type these slash commands for direct workspace control:
|
|
44
|
-
|
|
45
|
-
- **`/undo`**: Reverts the last assistant action, restoring files to their exact "before" state and rolling back messages.
|
|
46
|
-
- **`/redo`**: Re-applies the last undone turn, re-writing files and restoring conversation logs.
|
|
47
|
-
- **`/rewind <turn_index>`**: Rolls back the entire session to a specific conversation turn.
|
|
48
|
-
- **`/doctor`**: Run diagnostics on environment variables, Python version, dependencies, API model connections, and MCP server status.
|
|
49
|
-
- **`/report`**: Generates a support bundle under `.utim_tmp/report_bundle.zip` (automatically redacts secrets, passwords, or personal names/files).
|
|
50
|
-
- **`/reset`**: Wipes the current chat history without deleting persistent local vector memory.
|
|
51
|
-
|
|
52
|
-
---
|
|
53
|
-
|
|
54
|
-
## 🧠 Architecture & How it Works
|
|
55
|
-
|
|
56
|
-
1. **Local Memory (`.utim/memory.json` & ChromaDB)**:
|
|
57
|
-
- Global user preferences, rules, and facts are synced to a semantic vector database (`.utim_tmp/vector_db`).
|
|
58
|
-
- Relevant memories are dynamically fetched via semantic similarity (RAG) and injected into the system prompt context, preventing prompt bloating.
|
|
59
|
-
2. **Undo/Redo Stack & Session State**:
|
|
60
|
-
- Every file change (writes, batch string edits, moves, deletions) computes a diff snapshot.
|
|
61
|
-
- The entire stack is serialized dynamically to `.utim/session_state.json`. You can close your shell, shut down your computer, and resume later with intact rollback features.
|
|
62
|
-
3. **Workspace Boundary & Safety Controls**:
|
|
63
|
-
- Prior to writing files, UTIM performs **Pre-Commit Syntax Checks** (AST compilation for Python, JSON loads, JS/TS checks).
|
|
64
|
-
- If tests are available (`pytest`, `npm test`, etc.), UTIM runs them in a background **Regression testing loop**, prompting the model to self-heal code errors if assertions fail.
|
|
65
|
-
- **Interactive TUI mode** (`utim`): every file mutation shows an interactive diff dialog; the developer accepts, edits, or rejects individual hunks before they are applied.
|
|
66
|
-
- **CLI task mode** (`utim task`): when running in a real terminal, destructive operations (`rm`, package installs, `>` redirects) prompt for `y/n` confirmation. When stdin is piped/non-interactive all edits are auto-accepted.
|
|
67
|
-
- **Sandbox mode** (`utim --sandbox`): classifies every terminal command as safe or risky and blocks risky commands until explicitly approved.
|
|
68
|
-
|
|
69
|
-
---
|
|
70
|
-
|
|
71
|
-
## ⚠️ What this Tool Can and Cannot Do
|
|
72
|
-
|
|
73
|
-
### Can Do:
|
|
74
|
-
- Read, write, and patch codebases safely.
|
|
75
|
-
- Install and coordinate custom MCP (Model Context Protocol) servers.
|
|
76
|
-
- Self-heal syntax and test errors before files are written.
|
|
77
|
-
- Revert any file modification instantly.
|
|
78
|
-
|
|
79
|
-
### Cannot Do:
|
|
80
|
-
- **No Remote Code Execution**: Runs locally on your machine.
|
|
81
|
-
- **Unverified Sensitive Reads Blocked**: Reading files or memory matching sensitive keys (like passwords, secret codes, or personal data) is blocked unless verified via your configured verification code.
|
|
82
|
-
- **No Auto-Deletions**: Any command that deletes files outside the working directory is blocked automatically.
|
|
1
|
+
# UTIM CLI Agent: Enterprise Coder Assistant
|
|
2
|
+
|
|
3
|
+
UTIM is an agentic developer CLI assistant designed to automate coding tasks directly inside your local terminal, featuring robust safety controls, self-healing quality gates, and local-first semantic memory.
|
|
4
|
+
|
|
5
|
+
---
|
|
6
|
+
|
|
7
|
+
## 🚀 Quick Start
|
|
8
|
+
|
|
9
|
+
### 1. Installation
|
|
10
|
+
Install the package from your local source directory:
|
|
11
|
+
|
|
12
|
+
```bash
|
|
13
|
+
# Basic installation
|
|
14
|
+
pip install .
|
|
15
|
+
|
|
16
|
+
# Recommended: Full installation (includes semantic vector RAG & web search)
|
|
17
|
+
pip install ".[full]"
|
|
18
|
+
```
|
|
19
|
+
|
|
20
|
+
### 2. Provider & Model Configuration
|
|
21
|
+
Start the chat session:
|
|
22
|
+
```bash
|
|
23
|
+
utim
|
|
24
|
+
```
|
|
25
|
+
On first run, UTIM checks for `.utim/config.json`. If it does not exist, it runs a provider and model configuration wizard. You can configure:
|
|
26
|
+
- **Default Providers**: OpenRouter, OpenAI, custom servers, etc.
|
|
27
|
+
- **Model Picker**: Press `Ctrl+M` in the chat terminal at any time to configure, pick, add, or delete LLMs.
|
|
28
|
+
|
|
29
|
+
---
|
|
30
|
+
|
|
31
|
+
## ⚡ Main CLI Commands
|
|
32
|
+
|
|
33
|
+
- **`utim`**: Starts the interactive chat terminal (TUI).
|
|
34
|
+
- **`utim task "<prompt>"`**: Executes a single task autonomously from the command line and exits. In an interactive terminal (`stdin` is a TTY) file writes and commands prompt for confirmation; when piped/scripted they run in auto-accept mode.
|
|
35
|
+
- **`utim --dry-run`**: Starts the session in **Dry-Run Mode** (all code modifications and shell commands are simulated, not written/executed).
|
|
36
|
+
- **`utim --sandbox`**: Runs all mutating shell command proposals in the intelligent local sandbox (untrusted commands will block until approved).
|
|
37
|
+
- **`utim doctor`** / **`utim init`** / **`utim reset`**: Administrative commands for state diagnosis, initialization, and factory resets.
|
|
38
|
+
|
|
39
|
+
---
|
|
40
|
+
|
|
41
|
+
## 🛠️ In-Chat Slash Commands
|
|
42
|
+
|
|
43
|
+
Inside the interactive chat terminal, type these slash commands for direct workspace control:
|
|
44
|
+
|
|
45
|
+
- **`/undo`**: Reverts the last assistant action, restoring files to their exact "before" state and rolling back messages.
|
|
46
|
+
- **`/redo`**: Re-applies the last undone turn, re-writing files and restoring conversation logs.
|
|
47
|
+
- **`/rewind <turn_index>`**: Rolls back the entire session to a specific conversation turn.
|
|
48
|
+
- **`/doctor`**: Run diagnostics on environment variables, Python version, dependencies, API model connections, and MCP server status.
|
|
49
|
+
- **`/report`**: Generates a support bundle under `.utim_tmp/report_bundle.zip` (automatically redacts secrets, passwords, or personal names/files).
|
|
50
|
+
- **`/reset`**: Wipes the current chat history without deleting persistent local vector memory.
|
|
51
|
+
|
|
52
|
+
---
|
|
53
|
+
|
|
54
|
+
## 🧠 Architecture & How it Works
|
|
55
|
+
|
|
56
|
+
1. **Local Memory (`.utim/memory.json` & ChromaDB)**:
|
|
57
|
+
- Global user preferences, rules, and facts are synced to a semantic vector database (`.utim_tmp/vector_db`).
|
|
58
|
+
- Relevant memories are dynamically fetched via semantic similarity (RAG) and injected into the system prompt context, preventing prompt bloating.
|
|
59
|
+
2. **Undo/Redo Stack & Session State**:
|
|
60
|
+
- Every file change (writes, batch string edits, moves, deletions) computes a diff snapshot.
|
|
61
|
+
- The entire stack is serialized dynamically to `.utim/session_state.json`. You can close your shell, shut down your computer, and resume later with intact rollback features.
|
|
62
|
+
3. **Workspace Boundary & Safety Controls**:
|
|
63
|
+
- Prior to writing files, UTIM performs **Pre-Commit Syntax Checks** (AST compilation for Python, JSON loads, JS/TS checks).
|
|
64
|
+
- If tests are available (`pytest`, `npm test`, etc.), UTIM runs them in a background **Regression testing loop**, prompting the model to self-heal code errors if assertions fail.
|
|
65
|
+
- **Interactive TUI mode** (`utim`): every file mutation shows an interactive diff dialog; the developer accepts, edits, or rejects individual hunks before they are applied.
|
|
66
|
+
- **CLI task mode** (`utim task`): when running in a real terminal, destructive operations (`rm`, package installs, `>` redirects) prompt for `y/n` confirmation. When stdin is piped/non-interactive all edits are auto-accepted.
|
|
67
|
+
- **Sandbox mode** (`utim --sandbox`): classifies every terminal command as safe or risky and blocks risky commands until explicitly approved.
|
|
68
|
+
|
|
69
|
+
---
|
|
70
|
+
|
|
71
|
+
## ⚠️ What this Tool Can and Cannot Do
|
|
72
|
+
|
|
73
|
+
### Can Do:
|
|
74
|
+
- Read, write, and patch codebases safely.
|
|
75
|
+
- Install and coordinate custom MCP (Model Context Protocol) servers.
|
|
76
|
+
- Self-heal syntax and test errors before files are written.
|
|
77
|
+
- Revert any file modification instantly.
|
|
78
|
+
|
|
79
|
+
### Cannot Do:
|
|
80
|
+
- **No Remote Code Execution**: Runs locally on your machine.
|
|
81
|
+
- **Unverified Sensitive Reads Blocked**: Reading files or memory matching sensitive keys (like passwords, secret codes, or personal data) is blocked unless verified via your configured verification code.
|
|
82
|
+
- **No Auto-Deletions**: Any command that deletes files outside the working directory is blocked automatically.
|
|
83
|
+
|
|
84
|
+
---
|
|
85
|
+
|
|
86
|
+
## 📋 Compatibility Matrix
|
|
87
|
+
|
|
88
|
+
| Component | Minimum Version | Recommended Version | Notes |
|
|
89
|
+
|-----------|-----------------|---------------------|-------|
|
|
90
|
+
| **Python** | 3.10 | 3.11+ | 3.10 required for asyncio features |
|
|
91
|
+
| **pip** | 21.0 | 23.0+ | For modern dependency resolution |
|
|
92
|
+
| **Node.js** | 18.0 | 20.0+ | Only if using MCP servers |
|
|
93
|
+
| **Operating System** | | | |
|
|
94
|
+
| - Windows | 10 (1903+) | 11 | PowerShell or CMD required |
|
|
95
|
+
| - macOS | 11 (Big Sur) | 14 (Sonoma) | Terminal.app or iTerm2 |
|
|
96
|
+
| - Linux | Ubuntu 20.04 | Ubuntu 22.04+ | Any modern distro with bash |
|
|
97
|
+
| **Terminals** | | | |
|
|
98
|
+
| - Windows Terminal | 1.0 | 1.19+ | Full color support |
|
|
99
|
+
| - iTerm2 | 3.4+ | 3.5+ | Recommended on macOS |
|
|
100
|
+
| - VS Code Terminal | 1.70+ | 1.85+ | Full support |
|
|
101
|
+
| **Required Tools** | | | |
|
|
102
|
+
| - git | 2.30 | 2.40+ | For undo/redo and diffs |
|
|
103
|
+
| - Docker | 20.10 | 24.0+ | Optional, for sandbox mode |
|
|
104
|
+
|
|
105
|
+
### Supported LLM Providers
|
|
106
|
+
|
|
107
|
+
| Provider | API Type | Models Supported |
|
|
108
|
+
|----------|----------|------------------|
|
|
109
|
+
| OpenRouter | REST | Claude, GPT, Gemini, Mistral, etc. |
|
|
110
|
+
| OpenAI | REST | GPT-4o, GPT-4 Turbo, GPT-3.5 |
|
|
111
|
+
| Anthropic | REST | Claude 3.5, 3, 2 |
|
|
112
|
+
| Google AI | REST | Gemini 1.5 Pro/Flash |
|
|
113
|
+
| Azure OpenAI | REST | GPT-4, GPT-3.5 |
|
|
114
|
+
| Ollama | Local | Llama 3, Mistral, CodeLlama |
|
|
115
|
+
| Custom Server | REST | Any OpenAI-compatible API |
|
|
116
|
+
|
|
117
|
+
---
|
|
118
|
+
|
|
119
|
+
## 📄 Related Documentation
|
|
120
|
+
|
|
121
|
+
- [📖 Full Documentation](https://utim.dev/docs)
|
|
122
|
+
- [🔒 Security Policy](SECURITY.md)
|
|
123
|
+
- [📊 SLA & Support](SLA.md)
|
|
124
|
+
- [🛡️ Privacy Policy](landing/src/docs_md/privacy.md)
|