> ## Documentation Index
> Fetch the complete documentation index at: https://docs.kisuke.dev/llms.txt
> Use this file to discover all available pages before exploring further.

# Coding with AI

> AI-powered coding assistant integrated into Kisuke

## Overview

Kisuke includes an integrated AI coding assistant. Get debugging help, code generation, and documentation without leaving the app — from any device. Connect your own account from any supported provider.

Kisuke is **provider agnostic** — all skills, slash commands, and features work across Anthropic, OpenAI, Google, and every other supported provider.

## Getting Started

1. Open Kisuke
2. Tap **Chat**
3. Sign in with your AI provider
4. Start coding with AI

## Supported Providers

### Official (First-Party)

<CardGroup cols={3}>
  <Card title="Anthropic" icon={<svg viewBox="0 0 24 24" xmlns="http://www.w3.org/2000/svg"><path fill="currentColor" d="M17.3041 3.541h-3.6718l6.696 16.918H24Zm-10.6082 0L0 20.459h3.7442l1.3693-3.5527h7.0052l1.3693 3.5528h3.7442L10.5363 3.5409Zm-.3712 10.2232 2.2914-5.9456 2.2914 5.9456Z"/></svg>}>
    Claude models — OAuth or API key
  </Card>

  <Card title="OpenAI" icon={<svg viewBox="0 0 16 16" xmlns="http://www.w3.org/2000/svg"><path fill="currentColor" d="M14.949 6.547a3.94 3.94 0 0 0-.348-3.273 4.11 4.11 0 0 0-4.4-1.934A4.1 4.1 0 0 0 8.423.2 4.15 4.15 0 0 0 6.305.086a4.1 4.1 0 0 0-1.891.948 4.04 4.04 0 0 0-1.158 1.753 4.1 4.1 0 0 0-1.563.679A4 4 0 0 0 .554 4.72a3.99 3.99 0 0 0 .502 4.731 3.94 3.94 0 0 0 .346 3.274 4.11 4.11 0 0 0 4.402 1.933c.382.425.852.764 1.377.995.526.231 1.095.35 1.67.346 1.78.002 3.358-1.132 3.901-2.804a4.1 4.1 0 0 0 1.563-.68 4 4 0 0 0 1.14-1.253 3.99 3.99 0 0 0-.506-4.716m-6.097 8.406a3.05 3.05 0 0 1-1.945-.694l.096-.054 3.23-1.838a.53.53 0 0 0 .265-.455v-4.49l1.366.778q.02.011.025.035v3.722c-.003 1.653-1.361 2.992-3.037 2.996m-6.53-2.75a2.95 2.95 0 0 1-.36-2.01l.095.057L5.29 12.09a.53.53 0 0 0 .527 0l3.949-2.246v1.555a.05.05 0 0 1-.022.041L6.473 13.3c-1.454.826-3.311.335-4.15-1.098m-.85-6.94A3.02 3.02 0 0 1 3.07 3.949v3.785a.51.51 0 0 0 .262.451l3.93 2.237-1.366.779a.05.05 0 0 1-.048 0L2.585 9.342a2.98 2.98 0 0 1-1.113-4.094zm11.216 2.571L8.747 5.576l1.362-.776a.05.05 0 0 1 .048 0l3.265 1.86a3 3 0 0 1 1.173 1.207 2.96 2.96 0 0 1-.27 3.2 3.05 3.05 0 0 1-1.36.997V8.279a.52.52 0 0 0-.276-.445m1.36-2.015-.097-.057-3.226-1.855a.53.53 0 0 0-.53 0L6.249 6.153V4.598a.04.04 0 0 1 .019-.04L9.533 2.7a3.07 3.07 0 0 1 3.257.139c.474.325.843.778 1.066 1.303.223.526.289 1.103.191 1.664zM5.503 8.575 4.139 7.8a.05.05 0 0 1-.026-.037V4.049c0-.57.166-1.127.476-1.607s.752-.864 1.275-1.105a3.08 3.08 0 0 1 3.234.41l-.096.054-3.23 1.838a.53.53 0 0 0-.265.455zm.742-1.577 1.758-1 1.762 1v2l-1.755 1-1.762-1z"/></svg>}>
    GPT models — OAuth or API key
  </Card>

  <Card title="Google" icon={<svg viewBox="0 0 16 16" xmlns="http://www.w3.org/2000/svg"><path fill="currentColor" d="M15.545 6.558a9.4 9.4 0 0 1 .139 1.626c0 2.434-.87 4.492-2.384 5.885h.002C11.978 15.292 10.158 16 8 16A8 8 0 1 1 8 0a7.7 7.7 0 0 1 5.352 2.082l-2.284 2.284A4.35 4.35 0 0 0 8 3.166c-2.087 0-3.86 1.408-4.492 3.304a4.8 4.8 0 0 0 0 3.063h.003c.635 1.893 2.405 3.301 4.492 3.301 1.078 0 2.004-.276 2.722-.764h-.003a3.7 3.7 0 0 0 1.599-2.431H8v-3.08z"/></svg>}>
    Gemini models — OAuth or API key
  </Card>

  <Card title="xAI" icon={<svg viewBox="0 0 727.27 778.68" xmlns="http://www.w3.org/2000/svg"><polygon fill="currentColor" transform="translate(-134,-113.32)" points="508.67 574.07 761.27 213.32 639.19 213.32 447.64 486.9"/><polygon fill="currentColor" transform="translate(-134,-113.32)" points="356.08 792 417.12 704.83 356.08 617.66 234 792"/><polygon fill="currentColor" transform="translate(-134,-113.32)" points="508.67 792 630.75 792 356.08 399.72 234 399.72"/><polygon fill="currentColor" transform="translate(-134,-113.32)" points="761.27 256.91 661.27 399.72 671.27 792 751.27 792"/></svg>}>
    Grok models — API key
  </Card>

  <Card title="zAI" icon={<svg viewBox="0 0 30 30" xmlns="http://www.w3.org/2000/svg"><path fill="currentColor" d="M15.47,7.1l-1.3,1.85c-0.2,0.29-0.54,0.47-0.9,0.47h-7.1V7.09C6.16,7.1,15.47,7.1,15.47,7.1z"/><polygon fill="currentColor" points="24.3,7.1 13.14,22.91 5.7,22.91 16.86,7.1"/><path fill="currentColor" d="M14.53,22.91l1.31-1.86c0.2-0.29,0.54-0.47,0.9-0.47h7.09v2.33H14.53z"/></svg>}>
    zAI models — API key
  </Card>
</CardGroup>

### Third-Party

Connect via API key to any of these providers:

Azure, Cerebras, DeepSeek, Groq, Moonshot, OpenRouter, Qwen, Together AI, Vercel

## Skills

Type `/` in the chat input to bring up the skills picker. All skills work with Anthropic, OpenAI, Google, and every other supported provider.

### Conversation

| Command    | Description                                                |
| ---------- | ---------------------------------------------------------- |
| `/clear`   | Clear the current conversation and start fresh             |
| `/compact` | Summarize and compress the conversation to free up context |
| `/rename`  | Rename the current conversation                            |
| `/resume`  | Resume a previous conversation                             |
| `/rewind`  | Rewind the conversation to a previous state                |

### Agentic Modes

| Command  | Description                                                            |
| -------- | ---------------------------------------------------------------------- |
| `/agent` | Switch to agent mode — the AI takes multi-step actions autonomously    |
| `/auto`  | Let the AI decide the best approach automatically                      |
| `/plan`  | Ask the AI to plan before acting, so you can review the approach first |

### Code & Review

| Command     | Description                                            |
| ----------- | ------------------------------------------------------ |
| `/add-dir`  | Add a directory to the AI's context                    |
| `/review`   | Request a code review of the current file or selection |
| `/security` | Run a security-focused review of your code             |

### Other

| Command   | Description                                                  |
| --------- | ------------------------------------------------------------ |
| `/skills` | List all available skills                                    |
| `/todo`   | View and manage todos the AI has tracked during your session |

### Usage & Stats

| Command        | Description                               |
| -------------- | ----------------------------------------- |
| `/usage`       | View token usage for the current session  |
| `/stats`       | View detailed session statistics          |
| `/extra-usage` | View extended usage and quota information |

## Chat Interface

### @ File Mentions

Type `@` in the chat input to bring up a file selection menu. Tap any file from the list to insert it as a mention — this gives the AI direct context from that file without needing to paste its contents manually.

### Code Blocks

AI responses include syntax-highlighted code blocks. Tap to:

* **Copy** - Copy code to clipboard
* **Insert** - Insert into current file
* **Run** - Execute in terminal (if appropriate)

### Messages

* **You** - Your messages appear on the right
* **AI** - Responses appear on the left with syntax highlighting

### Conversation History

Previous conversations are saved per Session. Access them by:

1. Tap the history icon in chat
2. Select a previous conversation
3. Continue where you left off

## Keyboard Shortcuts

| Shortcut      | Action                                     |
| ------------- | ------------------------------------------ |
| `Shift + Tab` | Toggle between chat input and editor focus |
| `Ctrl + S`    | Stash the current prompt to continue later |
| `Alt + V`     | Attach an image (Desktop)                  |

## GitHub Actions

Kisuke's AI chat integrates with GitHub Actions — trigger or monitor workflows directly from the chat interface without leaving your session.

## Features

### Code Context

When you're in a Session, the AI has access to:

* Your project files
* Recent terminal output
* Previous conversation history

This helps the AI give relevant, contextual answers specific to your codebase.

### Code Generation

Ask the AI to write code:

```
"Write a Python function to parse CSV files"
"Create a React component for a login form"
"Generate SQL to find duplicate records"
```

### Debugging Help

Share errors and get solutions:

```
"Why am I getting this error: [paste error]"
"My API call returns 403, what could be wrong?"
"This function returns undefined, help me debug"
```

### Code Explanation

Understand unfamiliar code:

```
"Explain what this regex does: ^[a-zA-Z0-9+_.-]+@[a-zA-Z0-9.-]+$"
"Walk me through this algorithm"
"What does this SQL query do?"
```

### Documentation

Generate docs for your code:

```
"Add JSDoc comments to this function"
"Write a README for this project"
"Document this API endpoint"
```

## Tips

### Be Specific

More detail = better answers:

```
❌ "Fix my code"
✅ "My React useEffect runs infinitely. Here's my component: [code]. How do I fix the dependency array?"
```

### Share Context

Include relevant files or errors:

```
"Looking at src/utils/parser.js, can you optimize the parse function?"
```

### Follow Up

Continue the conversation to refine answers:

```
"That works, but can you make it handle null values?"
```

## Privacy

* Conversations are sent to your chosen AI provider
* Kisuke doesn't store your chat content on our servers
* Your API credentials stay on your device