Aide

approved

by doko

AI writing assistant powered by OpenAI-compatible APIs. - This plugin has not been manually reviewed by Obsidian staff.

137 downloadsUpdated 1mo agoMIT

Aide

AI writing assistant for Obsidian. OpenAI-compatible, streaming, tool calling.

Features

  • Slash trigger — type / (or \) to open a floating prompt
  • Inline actions — improve, shorten, summarize, translate, fix grammar, and more
  • Custom prompt — type any question or instruction in the input bar
  • Streaming — response appears character by character in real-time
  • Web search — AI can search Google, Brave, or a custom API for current info
  • Web fetch — AI can read any URL and summarize its content
  • SVG generation — AI can create diagrams and charts, save to assets/, and insert ![](...) into your note
  • Reasoning effort — supports OpenAI o-series reasoning_effort (low/medium/high)
  • Multi-provider — works with any OpenAI-compatible API (OpenAI, Anthropic via proxy, local Ollama, etc.)

Installation

From Obsidian Community Plugins

Search Aide in Settings → Community plugins → Browse.

Manual

  1. Download main.js, manifest.json, styles.css from releases
  2. Copy to <vault>/.obsidian/plugins/aide/
  3. Enable in Settings → Community plugins

Usage

  1. Open any note and type / (or backslash, configurable)
  2. Select an action from the list, or type a custom prompt and press Enter
  3. Watch the AI generate or edit your text in real-time

Configuration

SettingDescription
API Base URLOpenAI-compatible endpoint
ModelModel identifier (e.g. gpt-4o-mini, o3-mini, claude-3-opus)
Max tokensResponse length limit
TemperatureRandomness (0 = deterministic, 2 = very random)
Reasoning efforto-series only — how long the model thinks
Web searchEnable Google/Brave/Custom search via tool calling
SVG assets folderFolder (relative to vault root) for generated SVGs

Development

git clone https://github.com/doko89/obsidian-aide.git
cd obsidian-aide
npm install
npm run dev    # watch mode
npm run build  # production build

License

MIT

For plugin developers

Search results and similarity scores are powered by semantic analysis of your plugin's README. If your plugin isn't appearing for searches you'd expect, try updating your README to clearly describe your plugin's purpose, features, and use cases.