mirror of
https://github.com/ahmetildirim/obsidian-inscribe.git
synced 2026-07-22 05:44:10 +00:00
Update README to enhance setup instructions and add new images for provider configuration and profile assignments
This commit is contained in:
parent
fb5a86e11c
commit
1f162d0c2b
5 changed files with 26 additions and 42 deletions
68
README.md
68
README.md
|
|
@ -1,56 +1,40 @@
|
|||
# Obsidian Inscribe Plugin
|
||||
|
||||
Inscribe is an AI-powered inline autocompletion plugin for Obsidian. After you type a space, the plugin can suggest natural continuations of your text, helping you write more efficiently.
|
||||
Inscribe is an AI-powered inline autocompletion plugin for Obsidian. It will generate autocompletions as you type and you can accept or reject them with TAB.
|
||||
|
||||
## Features
|
||||
## Get started
|
||||
### 1. Setup a provider
|
||||
Inscribe supports following providers:
|
||||
- Ollama (local)
|
||||
- OpenAI
|
||||
- OpenAI compatible
|
||||
|
||||
### AI-Powered Completions
|
||||
Get intelligent, context-aware text suggestions as you write.
|
||||
More providers will be added if requested. Or better, feel free to create PR.
|
||||
|
||||
### Multiple AI Provider Support**:
|
||||
- OpenAI (GPT-4o, GPT-4o-mini)
|
||||
- Ollama (local models like Llama3, Mistral, Gemma)
|
||||
- OpenAI-compatible APIs
|
||||

|
||||
|
||||
### Customizable Profiles
|
||||
Create different completion profiles for various writing styles and needs
|
||||
### 2. Customize prompts
|
||||
You can set both system and user prompts to guide the AI's responses. For user prompt, template variables (mustache) can be used to provide context;
|
||||
- **{{pre_cursor}}**: text before cursor
|
||||
- **{{post_cursor}}**: text after cursor
|
||||
- **{{active_sentence}}**: the current sentence
|
||||
- **{{last_line}}**: last line of the active note
|
||||
|
||||
### Per-Path Configuration
|
||||
Set different profiles for different folders in your vault
|
||||

|
||||
|
||||
### Adjustable Parameters
|
||||
- Customize temperature to control creativity level
|
||||
- Fine-tune delay settings
|
||||
- Edit system and user prompts
|
||||
### 3. Per-Path profile assignments (Optional)
|
||||
You can create multiple profiles and assign them to specific note paths.
|
||||
For example, you might want a more formal profile for research notes and a more creative one for writing fiction. Corresponding profile will be automatically activated when you open a note matching the assigned path.
|
||||
|
||||
## How to Use
|
||||

|
||||
|
||||
1. **Installation**: Install the plugin through Obsidian's Community Plugins browser or manually
|
||||
2. **Configuration**:
|
||||
- Go to Settings → Inscribe
|
||||
- Enable the plugin and configure at least one AI provider
|
||||
- Set up your API keys or connection details for your chosen provider(s)
|
||||
3. **Create Profiles** (optional):
|
||||
- Create different profiles with specific AI models and settings
|
||||
- Assign profiles to specific paths in your vault
|
||||
4. **Start Writing**:
|
||||
- Begin typing in any note
|
||||
- After typing a space, the plugin will suggest completions
|
||||
- Press Tab to accept a suggestion or continue typing to ignore it
|
||||
5. **Toggle On/Off**:
|
||||
- Use the status bar icon to quickly enable/disable completions
|
||||
- Configure path-specific settings to control where completions appear
|
||||
|
||||
## Setup Tips
|
||||
## Tips
|
||||
### What provider should you use?
|
||||
Given the high frequency of API calls, using Ollama is highly recommended to reduce costs and improve response times. And for Ollama models, consider using a model bigger than 8B params. For now, `gemma3:12b` is a good starting point.
|
||||
|
||||
- For OpenAI: Enter your API key in the settings
|
||||
- For Ollama: Make sure your Ollama server is running (default: http://localhost:11434)
|
||||
- Adjust temperature for more conservative (lower) or creative (higher) completions
|
||||
- Customize prompts to better match your writing style and needs
|
||||
## Disable completions using path config
|
||||
Using per-path profile config, you can enable completions only for specific paths. This way, you can avoid annoying interruptions in notes where autocompletion isn't desired.
|
||||
|
||||
---
|
||||
|
||||
Made with ❤️ by [Ahmet Ildirim](https://github.com/ahmetildirim)
|
||||
|
||||
If you find this plugin useful, consider [buying me a coffee](https://buymeacoffee.com/ahmetildirim)
|
||||

|
||||
|
||||
|
|
|
|||
BIN
doc/images/customize-prompts.png
Normal file
BIN
doc/images/customize-prompts.png
Normal file
Binary file not shown.
|
After Width: | Height: | Size: 242 KiB |
BIN
doc/images/disable-completions.png
Normal file
BIN
doc/images/disable-completions.png
Normal file
Binary file not shown.
|
After Width: | Height: | Size: 156 KiB |
BIN
doc/images/per-path-profile-assignments.png
Normal file
BIN
doc/images/per-path-profile-assignments.png
Normal file
Binary file not shown.
|
After Width: | Height: | Size: 158 KiB |
BIN
doc/images/setup-provider.png
Normal file
BIN
doc/images/setup-provider.png
Normal file
Binary file not shown.
|
After Width: | Height: | Size: 88 KiB |
Loading…
Reference in a new issue