https://github.com/hcoz/sem-cli
A terminal app that works with human language commands.
https://github.com/hcoz/sem-cli
cli client command-line human-language-commands natural-language-processing nlp nodejs wit
Last synced: about 2 months ago
JSON representation
A terminal app that works with human language commands.
- Host: GitHub
- URL: https://github.com/hcoz/sem-cli
- Owner: hcoz
- License: mit
- Created: 2018-12-03T05:47:59.000Z (over 7 years ago)
- Default Branch: master
- Last Pushed: 2020-11-11T14:10:28.000Z (over 5 years ago)
- Last Synced: 2026-04-29T06:38:40.771Z (about 2 months ago)
- Topics: cli, client, command-line, human-language-commands, natural-language-processing, nlp, nodejs, wit
- Language: JavaScript
- Homepage: https://link.medium.com/DhlnH5r0ZZ
- Size: 18.6 KB
- Stars: 3
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# sem-cli
* This is a terminal app which works with human language commands.
* It is developed in Node.js without any framework or external library.
* This is the client side of the project. There is not so much logic in it. You can find the server side of it in there:
## Installation
You can install as a npm package. I recommend global install but it is up to you!
```bash
npm install sem-cli -g
```
## Commands
* **sem-look**: It is for showing the corresponding command for your message.
```bash
sem-look
```
You can use the all languages which are supported by Wit.ai
To check list of all languages, visit
Here is the example:
```bash
sem-look list files
```
You can also add arguments to your messages by using this notation `p=""`
Here is the example:
```bash
sem-look compare files p="a.txt b.txt"
```
It will return `Your command: "cmp a.txt b.txt" with danger level: "low"` for MacOS
* **sem-exec**: It is for executing the corresponding command for your message.
```bash
sem-exec
```
You can use the all languages which are supported by Wit.ai
To check list of all languages, visit
Here is the example:
```bash
sem-exec list files
```
You can also add arguments to your messages by using this notation `p=""`
Here is the example:
```bash
sem-exec list files p="subdir"
```
It will show the files under `/`
* **sem-suggest**: It is for suggesting a new (intent, command, dangerLevel) triple to extend the database
* **intent**: It is a short explanation for the purpose of this command. It will be the key of this triple
* **command**: It is the related command. (Please suggest a command which is working on your current operating system)
* **dangerLevel**: It shows how dangerous to run this command. It can be "low", "medium" or "high". ("high" dangerous commands will not be run before a client approval)
**The order of parameters is important. Please use same as the shown below.**
```bash
sem-suggest intent="" command="" dangerLevel=""
```
Here is the example:
```bash
sem-suggest intent="list_files" command="ls" dangerLevel="low"
```