anirudh0616/docright.nvim — explained in plain English
Analysis updated 2026-05-18
Get a quick documentation lookup for unfamiliar code without leaving Neovim
Ask follow-up questions about a specific part of an AI-generated explanation
Run AI-assisted code documentation fully locally without a remote API key
| anirudh0616/docright.nvim | jagdeep85/control4-kasa-cloud-driver | nklbdev/aseprite-stamp-brush | |
|---|---|---|---|
| Stars | 4 | 4 | 4 |
| Language | Lua | Lua | Lua |
| Setup difficulty | moderate | hard | easy |
| Complexity | 2/5 | 4/5 | 2/5 |
| Audience | developer | ops devops | designer |
Figures from each repo's GitHub metadata at analysis time.
Requires Ollama running locally with a pulled model such as qwen2.5-coder.
DocRight.nvim is a Neovim plugin that asks a locally running AI model for short documentation about whatever code is under your cursor or selected in visual mode. It is built for quick answers while you keep coding, not for open ended conversation. The plugin connects to Ollama, a tool for running AI language models on your own computer, and sends it the selected code along with a system prompt that tells the model to stay focused on programming topics such as languages, APIs, libraries, frameworks, and developer tools, and to refuse anything unrelated to programming. Responses are meant to read like a compact reference entry rather than a tutorial, listing the important parts of the code such as parameters, return values, side effects, or things to watch out for. To use it, you select code and press a key mapping to get documentation, or press another mapping to ask a follow up question about the last result. Inside the response window itself, you can select more text and press the same key to dig deeper into a specific part of the answer. The plugin also includes debug commands that show you the exact prompt being sent and the final merged configuration, which is useful if the responses are not what you expect. Setup requires Neovim 0.9 or later, the curl command line tool, and Ollama running locally with a model already pulled, such as qwen2.5-coder. Configuration options let you change the model, the Ollama endpoint, the response length, how many lines of surrounding code are sent for context, the size and position of the response window, and the key mappings used to trigger it. Everything runs locally with no remote API required, which the author lists as one of the plugin's main reasons for building it.
A Neovim plugin that asks a local Ollama AI model for short documentation about the code under your cursor.
Mainly Lua. The stack also includes Lua, Neovim, Ollama.
Setup difficulty is rated moderate, with roughly 30min to a first successful run.
Mainly developer.
This repo across BitVibe Labs
Verify against the repo before relying on details.