Skip to content

rodrigooler/claude-code-tldr

tldr — TL;DR mode for Claude Code

License: MIT Claude Code Plugin PRs welcome

A toggle that makes Claude Code answer ultra-short — every reply ≤ 5 lines, until you turn it off.

Claude is brilliant but loves to over-explain. tldr is a persistent ON/OFF mode: flip it on and you get just the result and the next step — no preamble, no recap, no walls of text. Flip it off and Claude goes back to normal.


Why

Reading long terminal output is tiring. Sometimes you just want the answer. tldr trades verbosity for signal on demand, without you having to nag for brevity every single turn.

It is smart about it: code you will paste/run and security warnings are never truncated — only the chatter around them is.

See it

Before (default):

Great question! Let me walk you through this. There are a few things to
consider here. First, the cache is stored in ~/.cache, and the reason it
grows is because... (12 more lines) ... so in summary, run `cache clear`.

After (tldr ON):

Run: cache clear
Cache lives in ~/.cache; it grows because nothing prunes it.

Install (2 lines)

In any Claude Code session:

/plugin marketplace add rodrigooler/claude-code-tldr
/plugin install tldr@claude-code-tldr

That's it. The first command registers this repo as a plugin marketplace; the second installs the tldr plugin.

Usage

Turn it on — type /tldr, or just say it:

Say this Effect
/tldr, "short mode", "be brief", "keep it short", "tl;dr" ON — replies become ≤ 5 lines
/tldr off, "back to normal", "normal mode" OFF — back to default

Claude confirms with one line (🔅 tldr ON …) and the mode persists across turns until you turn it off.

You can also invoke the skill explicitly as /tldr:tldr (Claude Code namespaces plugin skills as /<plugin>:<skill>). The natural-language triggers above work too, because the skill is model-invoked from its description.

What changes while ON

  • Max 3–5 lines. Result + next step only.
  • No filler ("great question", "let me explain…"), no recap.
  • Short bullets, one idea per line.
  • Code/commands: only the snippet that changes.
  • Big decision or risk? You get the verdict in ≤ 5 lines plus an offer: say expand for the full detail.

What it never shortens

  • Code you will paste or run — always complete and correct.
  • Security / data-loss warnings — clarity wins.

How it works

tldr is a single Claude Code Skill shipped as a plugin. There is no MCP server, no background process, no telemetry — just one SKILL.md of behavioral rules that Claude loads when you toggle the mode. See skills/tldr/SKILL.md.

Manual install (without the plugin system)

Prefer not to use plugins? Copy the skill into your personal skills folder:

git clone https://github.com/rodrigooler/claude-code-tldr
mkdir -p ~/.claude/skills/tldr
cp claude-code-tldr/skills/tldr/SKILL.md ~/.claude/skills/tldr/SKILL.md

Restart Claude Code and use /tldr as above.

Update / uninstall

/plugin install tldr@claude-code-tldr      # re-run to update to the latest version
/plugin uninstall tldr@claude-code-tldr     # remove it

Contributing

Issues and PRs are welcome — see CONTRIBUTING.md. The whole plugin is one Markdown file, so improving the prompt is a one-file change.

License

MIT © rodrigooler

About

TL;DR mode for Claude Code — a one-toggle plugin/skill that makes Claude answer ultra-short (under 5 lines) until you turn it off.

Topics

Resources

License

Code of conduct

Contributing

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors