gigachat support

.cursor/rules/ai-output-variants.mdc

@@ -0,0 +1,69 @@
+---
+description: 
+globs: 
+alwaysApply: false
+---
+Examples of correct responses. Use these program responses as a reference to understand that it's working correctly.
+
+1.
+```
+> npx ai my ip
+
+┌  AI Shell
+│
+◇  Your script:
+│
+│  curl ipinfo.io/ip
+│
+◇  Explanation:
+│
+│  Request to a public service that returns your IP address.
+│
+◆  Run this script?
+│  ● ✅ Yes (Lets go!)
+│  ○ 📝 Revise
+│  ○ ❌ Cancel
+└
+```
+
+2.
+```
+> npx ai 'what is the weather in New York'
+
+┌  AI Shell
+│
+◇  Your script:
+│
+│  curl wttr.in/NewYork
+│
+◇  Explanation:
+│
+│  Popular weather service that prints weather in a beautiful format with ASCII graphics.
+│
+◆  Run this script?
+│  ● ✅ Yes (Lets go!)
+│  ○ 📝 Revise
+│  ○ ❌ Cancel
+└
+```
+
+3.
+```
+> npx ai find all log files
+
+┌  AI Shell
+│
+◇  Your script:
+│
+│  find . -type f -name '*.log'
+│
+◇  Explanation:
+│
+│  Searches for log files in the current directory
+│
+◆  Run this script?
+│  ● ✅ Yes (Lets go!)
+│  ○ 📝 Revise
+│  ○ ❌ Cancel
+└
+```

.cursor/rules/change-build-run-repeat.mdc

@@ -0,0 +1,22 @@
+---
+description: 
+globs: 
+alwaysApply: true
+---
+## Development Instructions (Always Apply)
+
+1. Install dependencies ($1): `npm install` - only if a new dependency was used and/or an old one was removed.
+2. Build ($2): `npm run build` - if code was changed and needs to be rebuilt.
+3. Run ($3): `npx ai 'description of what needs to be done'` - for functionality testing (for understanding how it should work, see `.cursor/rules/ai-output-variants.mdc`).
+4. Fix, log, debug, change dependencies ($1), rebuild ($2), and run ($3) as needed until the task is completed.
+
+p.s. For convenience, you can run commands ^2 and ^3 together using `$1 && $2` at once.
+
+## Additional (if required)
+
+Log ONLY using `src/helpers/logger.ts` (methods). Do not use `console.log`. Code snippets for use:
+- Import: `import { logger } from './logger';`
+- Call: `logger.debug(message: string, data?: any);` (as well as `info`, `error`)
+- See logs here: `./logs/ai-shell.log`
+
+If you need to understand the project structure, see `.cursor/rules/project-files-description.mdc`.

.cursor/rules/project-files-description.mdc

@@ -0,0 +1,72 @@
+---
+description: 
+globs: 
+alwaysApply: false
+---
+## Project Structure
+
+1. **Main files:**
+   - `src/cli.ts` - Entry point of the CLI application, handles user commands
+   - `src/prompt.ts` - Responsible for the beautiful animated interface and response display
+
+2. **AI engines (`src/helpers/engines/`):**
+   - `engine-api.ts` - Abstract interface for all AI engines
+   - `engine-factory.ts` - Factory for creating AI engine instances
+   - `config-platform.ts` - Configuration for AI engines
+
+3. **Helper modules (`src/helpers/`):**
+   - `completion.ts` - **Key file!** Implementation of OpenAI API integration, response generation and processing (should be in `src/helpers/engines/`, but left here for better git-diff)
+   - `logger.ts` - Logging system with file rotation support and debug/info/error levels
+   - `config.ts` - Application settings and configuration management
+   - `error.ts` - Error handling and exceptions
+   - `i18n.ts` - Internationalization and interface translations
+   - `shell-history.ts` - Shell command history management
+   - `constants.ts` - Application constants
+   - `os-detect.ts` - Operating system and shell detection
+   - `stream-to-iterable.ts` - Utility for converting streams to iterable objects
+   - `stream-to-string.ts` - Utility for converting streams to strings
+   - `strip-regex-patterns.ts` - Utility for cleaning text from regex patterns
+   - `replace-all-polyfill.ts` - Polyfill for the replaceAll method
+
+4. **Commands (`src/commands/`):**
+   - `chat.ts` - Interactive AI chat implementation
+   - `config.ts` - Application configuration via CLI interface
+   - `update.ts` - Automatic application updates
+
+5. **Localization (`src/locales/`):**
+   - Translation files in JSON format for different languages:
+     - `en.json` - English
+     - `ru.json` - Russian  
+     - `de.json` - German
+     - `fr.json` - French
+     - `es.json` - Spanish
+     - `pt.json` - Portuguese
+     - `zh-Hans.json`, `zh-Hant.json` - Chinese (Simplified and Traditional)
+     - `jp.json` - Japanese
+     - `ko.json` - Korean
+     - `ar.json` - Arabic
+     - `tr.json` - Turkish
+     - `uk.json` - Ukrainian
+     - `vi.json` - Vietnamese
+     - `id.json` - Indonesian
+
+6. **Configuration files (project root):**
+   - `package.json` - Package description and npm dependencies
+   - `tsconfig.json` - TypeScript compiler settings
+   - `.eslintrc.cjs` - ESLint settings for code checking
+   - `.prettierrc.json` - Code formatting settings
+   - `.nvmrc` - Node.js version for the project
+
+7. **Documentation:**
+   - `README.md` - Main project documentation
+   - `README-zh-Hans.md` - Documentation in Chinese
+   - `CHANGELOG.md` - Version change list
+   - `CONTRIBUTING.md` - Contributor guide
+   - `GIGACHAT_TZ.md` - Technical specification for GigaChat integration
+   - `LICENSE` - Project license
+
+8. **Working directories:**
+   - `dist/` - Compiled files for production
+   - `logs/` - Application log files
+   - `node_modules/` - Installed npm dependencies
+   - `gigachat-js/` - Repository with source code of the module for working with GigaChat API

.gitignore

@@ -1,3 +1,5 @@
 node_modules
 .DS_Store
-dist
+dist
+gigachat-js/
+logs/

CHANGELOG.md

@@ -1,3 +1,12 @@
+## dev
+
+- **NEW ENGINE**: Added support for GigaChat
+- **BREAKING**: Renamed `MODEL` config to `OPENAI_MODEL` for clarity
+- **DEFAULT**: Changed default OpenAI model to `gpt-4.1-nano` (most cost-effective)
+- **FEATURE**: Added ability to change default config file path via `AI_SHELL_CONFIG_PATH` environment variable
+- **FEATURE**: Added Proxy PAC URL configuration support (`PROXY_PAC_URL`)
+- **FEATURE**: Added separate `ALL_PROXY` configuration for each AI engine (`OPENAI_ALLPROXY`, `GIGACHAT_ALLPROXY`)
+
 ## 1.0.12
 
 - Bug fixes

README-zh-Hans.md

@@ -182,7 +182,7 @@ ai config
 │  ○ OpenAI Key
 │  ○ OpenAI API Endpoint
 │  ○ Silent Mode
-│  ● Model (gpt-4o-mini)
+│  ● Model (gpt-4.1-nano)
 │  ○ Language
 │  ○ Cancel
 └

README.md

@@ -34,17 +34,36 @@
    npm install -g @builder.io/ai-shell
    ```
 
-2. Retrieve your API key from [OpenAI](https://platform.openai.com/account/api-keys)
+2. Choose and configure your AI engine:
+
+### Option A: OpenAI (default)
+
+1. Retrieve your API key from [OpenAI](https://platform.openai.com/account/api-keys)
 
    > Note: If you haven't already, you'll have to create an account and set up billing.
 
-3. Set the key so ai-shell can use it:
+2. Set the key so ai-shell can use it:
+
+   ```sh
+   ai config set OPENAI_KEY=<your openai token>
+   ```
+
+   > The default model is `gpt-4.1-nano` (most cost-effective). You can change it using `ai config set OPENAI_MODEL=<model_name>`
+
+### Option B: GigaChat
+
+1. Get your GigaChat API credentials from [AI Platform](https://developers.sber.ru/docs/ru/gigachat/individuals-quickstart)
+
+2. Configure GigaChat as your AI engine:
 
    ```sh
-   ai config set OPENAI_KEY=<your token>
+   ai config set AI_ENGINE=GigaChat
+   ai config set GIGACHAT_KEY=<your gigachat api key>
    ```
 
-   This will create a `.ai-shell` file in your home directory.
+---
+
+   This will create a `.ai-shell` file in your home directory. You can change the path to the config file by setting the `AI_SHELL_CONFIG_PATH` environment variable.
 
 ## Usage
 
@@ -63,15 +82,18 @@ Then you will get an output like this, where you can choose to run the suggested
 ```bash
 ◇  Your script:
 │
-│  find . -name "*.log"
+│ find . -name "*.log"
 │
 ◇  Explanation:
 │
-│  1. Searches for all files with the extension ".log" in the current directory and any subdirectories.
+│ 1. Search current directory and subdirectories  
+│ 2. Find files ending with ".log"
 │
 ◆  Run this script?
 │  ● ✅ Yes (Lets go!)
-│  ○ 📝 Revise
+│  ○ 📝 Edit
+│  ○ 🔁 Revise
+│  ○ 📋 Copy
 │  ○ ❌ Cancel
 └
 ```
@@ -127,14 +149,38 @@ or save the option as a preference using this command:
 ai config set SILENT_MODE=true
 ```
 
-### Custom API endpoint
+### Custom API endpoints
 
-You can custom OpenAI API endpoint to set OPENAI_API_ENDPOINT（default: `https://api.openai.com/v1`）
+You can customize API endpoints for both engines:
 
+**For OpenAI** (default: `https://api.openai.com/v1`):
 ```sh
 ai config set OPENAI_API_ENDPOINT=<your proxy endpoint>
 ```
 
+**For GigaChat**:
+```sh
+ai config set GIGACHAT_API_ENDPOINT=<your gigachat endpoint>
+```
+
+### Proxy Configuration
+
+The application supports advanced proxy settings for both engines:
+
+**ALL_PROXY settings** (separate for each engine):
+```sh
+# For OpenAI
+ai config set OPENAI_ALLPROXY=<your_proxy_url>
+
+# For GigaChat  
+ai config set GIGACHAT_ALLPROXY=<your_proxy_url>
+```
+
+**Proxy PAC URL** (common setting):
+```sh
+ai config set PROXY_PAC_URL=<your_pac_url>
+```
+
 ### Set Language
 
 ![Language UI](https://user-images.githubusercontent.com/1784873/235330029-0a3b394c-d797-41d6-8717-9a6b487f1ae8.gif)
@@ -178,10 +224,17 @@ To get an interactive UI like below:
 
 ```bash
 ◆  Set config:
-│  ○ OpenAI Key
-│  ○ OpenAI API Endpoint
+│  ● AI Engine (OpenAI)
+│  ○ [OpenAI] Key
+│  ○ [OpenAI] Model
+│  ○ [OpenAI] API Endpoint
+│  ○ [OpenAI] ALL_PROXY
+│  ○ [GigaChat] Key
+│  ○ [GigaChat] Model
+│  ○ [GigaChat] API Endpoint
+│  ○ [GigaChat] ALL_PROXY
+│  ○ [Common] Proxy PAC URL
 │  ○ Silent Mode
-│  ● Model (gpt-4o-mini)
 │  ○ Language
 │  ○ Cancel
 └
@@ -209,12 +262,29 @@ ai update
 
 ## Common Issues
 
-### 429 error
+### OpenAI Issues
+
+#### 429 error
 
 Some users are reporting a 429 from OpenAI. This is due to incorrect billing setup or excessive quota usage. Please follow [this guide](https://help.openai.com/en/articles/6891831-error-code-429-you-exceeded-your-current-quota-please-check-your-plan-and-billing-details) to fix it.
 
 You can activate billing at [this link](https://platform.openai.com/account/billing/overview). Make sure to add a payment method if not under an active grant from OpenAI.
 
+### GigaChat Issues
+
+#### Authentication errors
+
+If you encounter authentication issues with GigaChat:
+1. Verify your API key is correct
+2. Ensure your GigaChat account has proper access rights
+3. Check that your credentials haven't expired
+
+#### Rate limiting
+
+GigaChat has its own rate limiting. If you encounter 429 errors:
+1. Wait a few moments before retrying
+2. Check your usage quotas in the Developer Console
+
 ## Motivation
 
 I am not a bash wizard, and am dying for access to the copilot CLI, and got impatient.