AI-Code-Suggestions

AI Code Suggestions plugin for Code on the Go

Requires the AI Core addon and an AI backend. This plugin holds no model of its own. It asks AI Core for every completion, so install AI Core and a backend and choose it in Preferences → Configuration → Agent. Without them, this plugin shows no completions.

Inline ghost-text code completions. As you type, the plugin debounces, asks an LLM for a completion at the cursor, and shows it as dimmed inline text via IdeEditorService.showInlineSuggestion().

The dependency is at runtime only — there is no compile-time dependency on ai-core. This plugin resolves LlmInferenceService through the Code On The Go plugin manager (SharedServices), and binds lazily. Until AI Core is active it stays silent, then enables itself when the inference service appears.

Architecture

┌─────────────────────────────┐
│  ai-code-suggestions (this) │  ← content-change listener, debounce, ghost text
└──────────────┬──────────────┘
               │ SharedServices (runtime) → LlmInferenceService
               ▼
┌─────────────────────────────┐
│  ai-core                    │  ← routes to the selected backend plugin
└─────────────────────────────┘

Features

• Inline ghost-text completions while typing
• 800 ms debounce to reduce LLM load
• LRU cache to avoid redundant calls
• Language-aware prompts (Kotlin, Java, Python, …)
• Graceful degradation when ai-core isn't loaded yet (binds lazily)

Permissions

This plugin declares no plugin.permissions of its own — it only registers a content-change listener and renders ghost text. Reading the configured backend/model and any network access happen inside ai-core's process, which declares those permissions.

The code before the cursor (at most 500 characters) is sent to the configured backend to generate a completion: on-device for AI Agent Local, or over HTTPS to Google, Anthropic or your OpenAI-compatible server for Gemini, Claude or OpenAI. Choose the backend accordingly for sensitive code.

Building

Prerequisites: Android SDK (API 33+), JDK 17. Create local.properties with sdk.dir=.... No NDK or native toolchain.

cd plugins/AI-Code-Suggestions
./gradlew assemblePlugin          # release  -> build/plugin/ai-code-suggestions.cgp
./gradlew assemblePluginDebug     # debug variant

The build resolves plugin-api.jar from the repo-root ../../libs/.

Installation

1. Build and install ai-core first (see ../ai-core/README.md).
2. Build this plugin, install build/plugin/ai-code-suggestions.cgp via Code on the Go's Plugin Manager, and restart the IDE.
3. Install an AI backend and choose it in Preferences → Configuration → Agent.

Key classes

• CodeSuggestionsPlugin.kt — lifecycle, content-change listener, debounce
• SuggestionProvider.kt — prompt building, LLM call, LRU cache

License

GPL-3.0 — same as AndroidIDE / Code on the Go.