Wordwise for Claude Code
Underlines terms in Claude's replies and shows their meaning in a hover card.
Illustration. Actual results depend on the mod, terminal and environment.
Install
/plugin marketplace add arpxspace/wordwise-cc
/plugin install wordwise-cc@wordwise-cc
/reload-pluginsRun these commands one at a time in a Claude Code session.
claude plugin marketplace add arpxspace/wordwise-cc
claude plugin install wordwise-cc@wordwise-ccThen run /reload-plugins in your session, or start a new session.
Please install the Claude Code mod "Wordwise for Claude Code" (https://github.com/arpxspace/wordwise-cc). Before installing, read the Security and Requirements sections of its README and tell me what it reads and writes, which commands it runs, whether it uses the network, and which Claude Code version, OS and terminal it needs. If everything looks fine, run claude plugin marketplace add arpxspace/wordwise-cc and claude plugin install wordwise-cc@wordwise-cc, then remind me to run /reload-plugins in my session.Send this to Claude Code so it reads the author’s instructions before installing.
git clone https://github.com/arpxspace/wordwise-cc
cd wordwise-cc
claude --plugin-dir .Clone the repository and try it in a local session with --plugin-dir.
What it accesses
Based on the author’s documentation, checked when listed; use claude plugin validate to check for yourself.
Reads
YesReads Claude's reply text, the terminal viewport and hover state, and the askForGlossary setting.
Writes files
UnspecifiedNot stated by the author; the code shows no file writes, only screen rendering.
Runs commands
UnspecifiedNot stated by the author; the code shows no commands or processes.
Network
YesAdds the glossary rule to the system prompt sent with every model request; no other requests seen.
Introduction
Wordwise is a Claude Code mod inspired by Kindle's Word Wise. It adds a rule to the system prompt in every session so Claude starts each reply with a short glossary block (*term*: meaning between --- lines). In the fullscreen layout it hides that block, underlines each term in the reply, and shows the term's meaning in a rounded card when the mouse rests on it.
- Works only in the fullscreen layout (
"tui": "fullscreen"), the only layout that sees the mouse; elsewhere the reply is drawn normally, glossary included. - Code boxes, tables and quotes get no underlines. Only the screen changes; the saved reply keeps the glossary (ctrl+o shows it).
- Turn off
askForGlossaryin/configif your CLAUDE.md already asks for a glossary.
Note: the glossary rule goes into every model request and replies spend extra tokens on the block. The author says it was tested on Claude Code 2.1.296 and still calls code plugins early access; no minimum version is stated.
Who it is for
- People reading Claude replies in an unfamiliar field who want terms explained on hover
- People who want Claude to define jargon, acronyms and names it picks
Requirements
- Claude Code ≥ 2.1.296
- 2.1.296 is the version the author tested, not a stated minimum.
- Set "tui": "fullscreen" in ~/.claude/settings.json and start a new session.
- The README says plugins with code are still early access.