Skip to content

Commit 9bebcd8

Browse files
committed
Docs update #2!
1 parent 67d39f0 commit 9bebcd8

3 files changed

Lines changed: 40 additions & 181 deletions

File tree

CONTRIBUTING.md

Lines changed: 0 additions & 50 deletions
Original file line numberDiff line numberDiff line change
@@ -40,53 +40,3 @@ git checkout -b feature/your-feature
4040
git commit -m "feat: your feature"
4141
git push origin feature/your-feature
4242
```
43-
44-
## Code Standards
45-
46-
- **Memory**: Use CCObject retain/release properly
47-
- **Style**: camelCase variables, PascalCase classes
48-
- **Testing**: Test for crashes and memory leaks
49-
- **Geode**: Follow [docs.geode-sdk.org](https://docs.geode-sdk.org)
50-
51-
## Development
52-
53-
```bash
54-
geode build
55-
```
56-
57-
Test in GD with Geode installed.
58-
59-
## Help Wanted
60-
61-
- UI improvements
62-
- Localization
63-
- Mobile optimization
64-
- Statistics tracking
65-
66-
## Structure
67-
68-
```
69-
EditorAI/
70-
├── src/main.cpp # Main logic
71-
├── resources/ # Assets
72-
├── mod.json # Configuration
73-
└── CMakeLists.txt # Build config
74-
```
75-
76-
## Testing Checklist
77-
78-
- [ ] Compiles without errors
79-
- [ ] Loads in GD
80-
- [ ] No crashes or leaks
81-
- [ ] Tested on your platform
82-
- [ ] Docs updated
83-
84-
## Getting Help
85-
86-
- GitHub Issues
87-
- Geode Discord
88-
- [docs.geode-sdk.org](https://docs.geode-sdk.org)
89-
90-
---
91-
92-
**Be respectful, help others, focus on code quality.**

README.md

Lines changed: 21 additions & 94 deletions
Original file line numberDiff line numberDiff line change
@@ -2,105 +2,32 @@
22

33
AI-powered level generation for Geometry Dash. Describe your level in plain text and watch AI build it in the editor.
44

5-
## Features
6-
7-
- 🤖 **Multiple AI Providers** - Gemini (free), Claude, ChatGPT, or Ollama (local)
8-
- 📦 **1000+ Objects** - Full GD object library with auto-updates
9-
- 🎨 **Full Control** - Colors, groups, triggers, and advanced properties
10-
-**Fast & Stable** - Progressive creation with live progress
11-
- 🔒 **Private** - Keys stored locally, no telemetry
12-
135
## Quick Start
146

15-
1. Install from Geode mod browser
16-
2. Open editor → Click **AI** button (top-right)
17-
3. **For Ollama**: Download from [ollama.com](https://ollama.com), run `ollama run llama2`, skip API key
18-
4. **For Cloud AI**: Click lock icon → Enter API key from [ai.google.dev](https://ai.google.dev) (free)
19-
5. Type prompt and generate!
20-
21-
## Setup
22-
23-
### Ollama (Recommended - FREE & Fast)
24-
1. Download [ollama.com](https://ollama.com)
25-
2. Run `ollama run llama2`
26-
3. No API key needed
27-
4. **Benefits**: Free, unlimited, private, offline
28-
29-
### Cloud APIs
30-
- **Gemini** - [ai.google.dev](https://ai.google.dev) - Free (15 req/min)
31-
- **Claude** - [console.anthropic.com](https://console.anthropic.com) - Paid (~$0.001-0.02)
32-
- **ChatGPT** - [platform.openai.com](https://platform.openai.com) - Paid (~$0.001-0.03)
33-
34-
**Note**: Gemini is free but less capable. Claude and ChatGPT handle complex levels better.
35-
36-
## Example Prompts
37-
38-
```
39-
"Simple platforming section with basic jumps"
40-
"Medium difficulty ship corridor with portals"
41-
"Hard wave section with red blocks and yellow spikes"
42-
"Extreme demon timing section with groups 1 and 2"
43-
```
44-
45-
## Settings
46-
47-
**Settings → Mods → Editor AI:**
48-
49-
- **AI Provider** - gemini / claude / openai / ollama
50-
- **Model** - Choose from 9+ models
51-
- **Difficulty** - easy / medium / hard / extreme
52-
- **Style** - modern / retro / minimalist / decorated
53-
- **Length** - short / medium / long / xl / xxl
54-
- **Max Objects** - 10 to 1,000,000
55-
- **Features** - Colors, groups, triggers (toggle in settings)
56-
- **Rate Limiting** - Recommended ON (3s default)
57-
58-
## ⚠️ Known Issues
59-
60-
### Editor Collab Incompatibility
61-
**`Editor Collab` mod is NOT compatible with Editor AI.** They break each other. You must:
62-
- Disable Editor Collab before using Editor AI, OR
63-
- Uninstall Editor Collab
64-
65-
This is a known incompatibility and cannot be fixed without changes to both mods.
66-
67-
## Troubleshooting
68-
69-
**"API Key Required"**
70-
- Click lock icon and enter key
71-
- For Ollama: No key needed, ensure Ollama is running
72-
73-
**Objects Not Appearing**
74-
- Check max objects setting
75-
- Try smaller generation first
76-
- Verify level isn't locked
77-
78-
**Slow Generation**
79-
- Normal! Creates 20-50 objects/second
80-
- Watch progress indicator
81-
- Prevents crashes and ensures stability
82-
83-
## Platform Support
84-
85-
- ✅ Windows (Tested)
86-
- ✅ macOS (Tested)
87-
- ✅ Android (Tested)
7+
1. Download the [latest release](https://github.com/entity12208/editorai/releases/latest)
8+
2. Open the game and navigate to the Geode page
9+
3. Press the manual download button and select the file
10+
4. Set up the mod
8811

89-
## Links
12+
## AI Setup
9013

91-
- **GitHub**: [entity12208/EditorAI](https://github.com/entity12208/EditorAI)
92-
- **Geode Docs**: [docs.geode-sdk.org](https://docs.geode-sdk.org)
93-
- **Report Issues**: GitHub Issues or Geode Discord
14+
### Ollama
15+
1. Download and install [Ollama](https://ollama.com)
16+
2. Download the official model by running `ollama pull editorai:<model>`, where `model` can be either `deepseek` (more creative) or `qwen` (more powerful). Each is 5.2 GB.
17+
3. Ensure the setting `ollama model` is set to `editorai`.
18+
4. Change the other settings to your requirements.
9419

95-
## Credits
20+
### Gemini, Claude, ChatGPT
9621

97-
- **Developer**: entity12208
98-
- **Framework**: Geode SDK
99-
- **AI Providers**: Google, Anthropic, OpenAI, Ollama
22+
1. Get your API key for [Gemini](https://aistudio.google.com/u/0/api-keys), [Claude](https://console.anthropic.com), or [ChatGPT](https://platform.openai.com/api-keys).
23+
2. Set the setting `Model` to your perfered model used by your provider
24+
3. Once in the UI, click the lock icon.
25+
4. Paste your API key into the box and save it.
10026

101-
---
27+
## Usage
10228

103-
**Version**: 2.1.2
104-
**Game**: Geometry Dash 2.2074
105-
**Geode**: 4.10.0+
106-
**Status**: ✅ Stable
29+
1. Download and set up the mod.
30+
2. Open the editor of any level.
31+
3. Press the AI button located in the top right.
32+
4. Enable or disable `Clear Level`, and enter your prompt into the box.
33+
5. Click `Generate` and wait!

mod.json

Lines changed: 19 additions & 37 deletions
Original file line numberDiff line numberDiff line change
@@ -10,7 +10,7 @@
1010
"name": "Editor AI",
1111
"version": "v2.1.2",
1212
"developer": "entity12208",
13-
"description": "Generate Geometry Dash levels using AI! Supports Gemini, Claude, and ChatGPT with full trigger support (Alpha, Move, Toggle) and group IDs. Now available on all platforms!",
13+
"description": "Generate Geometry Dash levels using AI!",
1414
"settings": {
1515
"ai-provider": {
1616
"type": "string",
@@ -22,7 +22,7 @@
2222
"model": {
2323
"type": "string",
2424
"name": "Model",
25-
"description": "AI model to use",
25+
"description": "Non-Ollama AI model to use",
2626
"default": "gemini-2.5-flash",
2727
"one-of": [
2828
"gemini-2.5-flash",
@@ -65,6 +65,22 @@
6565
"min": 10,
6666
"max": 1000000
6767
},
68+
"ollama-section": {
69+
"type": "title",
70+
"name": "Ollama (Local AI)"
71+
},
72+
"ollama-url": {
73+
"type": "string",
74+
"name": "Ollama Server URL",
75+
"description": "URL of your local Ollama server",
76+
"default": "http://localhost:11434"
77+
},
78+
"ollama-model": {
79+
"type": "string",
80+
"name": "Ollama Model",
81+
"description": "Model name (auto-detected when Ollama is running)",
82+
"default": "editorai"
83+
},
6884
"rate-limiting-section": {
6985
"type": "title",
7086
"name": "Rate Limiting"
@@ -82,7 +98,7 @@
8298
"default": 3,
8399
"min": 1,
84100
"max": 60,
85-
"enableIf": "enable-rate-limiting"
101+
"enable-if": "enable-rate-limiting"
86102
},
87103
"features-section": {
88104
"type": "title",
@@ -115,42 +131,8 @@
115131
"name": "Advanced Level Settings",
116132
"description": "Allow AI to modify level settings like background/ground colors, start speed/mode, dual mode, etc. WARNING: This is experimental!",
117133
"default": false
118-
},
119-
"ollama-section": {
120-
"type": "title",
121-
"name": "Ollama (Local AI)"
122-
},
123-
"ollama-url": {
124-
"type": "string",
125-
"name": "Ollama Server URL",
126-
"description": "URL of your local Ollama server",
127-
"default": "http://localhost:11434"
128-
},
129-
"ollama-model": {
130-
"type": "string",
131-
"name": "Ollama Model",
132-
"description": "Model name (auto-detected when Ollama is running)",
133-
"default": "deepseek-v3.1:671b-cloud"
134-
},
135-
"api-key-info": {
136-
"type": "title",
137-
"name": "API Key Management"
138-
},
139-
"api-key-status": {
140-
"type": "string",
141-
"name": "Status",
142-
"description": "Current API key status",
143-
"default": "No API key saved",
144-
"enableIf": "false"
145134
}
146135
},
147-
"platforms": [
148-
"win",
149-
"android32",
150-
"android64",
151-
"mac",
152-
"ios"
153-
],
154136
"tags": [
155137
"editor",
156138
"utility",

0 commit comments

Comments
 (0)