Installation
-
Using the IDE built-in plugin system:
Settings/Preferences > Plugins > Marketplace > Search for "shire" > Install
-
Get from Marketplace:
-
Manually:
Download the latest release and install it manually using Settings/Preferences > Plugins > ⚙️ > Install plugin from disk...
Examples: https://github.com/shire-lang/shire-spring-java-demo
Usage
“Hello World” example
- create a new file, like
hi.shire
, with content:
hi
- Run file with
Shire
plugin, and you will see AI response result.
Prepare for running hi.shire...
Shire Script: /Volumes/source/ai/shire/hi.shire
Shire Script Compile output:
Used model: gpt-4o
hi
--------------------
Hello! How's it going?
Process finished with exit code 0
Example 2
- first create a new file, like
sample.shire
, with content:
Explain code /file:src/main/java/com/example/Controller.java
- Run file with
Shire
plugin, and you will see AI response result.
Use in IDE
Example:
---
name: "AutoTest"
description: "AutoTest"
actionLocation: ContextMenu
interaction: AppendCursor
when: { $fileName.contains(".java") && $filePath.contains("src/main/java") }
---
@ext-context.autotest
Write unit test for following ${context.language} code.
${context.frameworkContext}
/file:src/main/kotlin/com/phodal/blog/controller/UserController.kt
Config LLM
Current we support OpenAI, GLM, 零一万物, Moonshot AI, DeepSeek AI. You need to configure your API Token and Model in Settings
-> Tools
-> Shire
.
OpenAI
- LLM API Host: Empty
- ModelName: gpt-4o
- Engine Token: your key
GLM 示例
- LLM API Host: https://open.bigmodel.cn/api/paas/v4/chat/completions
- ModelName: glm-4
- Engine Token: your key
零一万物
- LLM API Host: https://api.lingyiwangwu.com/v1/chat/completions
- ModelName: yi-34b-chat
- Engine Token: your key
Moonshot AI
- LLM API Host: https://api.moonshot.cn/v1/chat/completions
- ModelName: moonshot-v1-8k
- Engine Token: your key
DeepSeek AI
- LLM API Host: https://api.deepseek.com/v1/chat/completions
- ModelName: deepseek-chat
- Engine Token: your key