agentleFS
naticorrea-03naticorrea-03/claude-code-skills/skills/plain-english/SKILL.md

plain-english

Skill4 starsChanged 5 months agoMITView on GitHub

Rewrite an explanation so the user actually wants to read it: short, plain, no jargon. Use the MOMENT they signal a reply was too long, too technical, or they just don't want to read it: "/plain", "in plain english", "english please", "shorter", "tldr", "too technical", "too long", "say that simpler", "dumb it down", "eli5", "what does that actually mean", "just tell me", "redo that", "rephrase that", or any sign they're bouncing off a wall of text or jargon. Default target is the last thing Claude said; the user can also point it at any other text (an email, a doc, something they paste). On-demand only, never auto-fires.

15 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use find-skills.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • VercelSkill

    find-skills

    Helps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express interest in extending capabilities. This skill should be used when the user is looking for functionality that might exist as an installable skill.

    Also in 3 projects that use this

The file

We split this skill into the rules it gives an agent

5 rules. 0 are common, 5 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 15 edited versions we read keep these.

  • 5

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Default target is the last thing you said.

    line 16
  2. Distinctive:

    Cut the jargon: If a technical word is genuinely load-bearing, drop one plain phrase next to it in parentheses.

    line 21
  3. Distinctive:

    Keep it human: No filler, no corporate buzzwords, no "it's not X, it's Y" constructions, short sentences.

    line 24
  4. Distinctive:

    Before: (eight lines walking through every step of how a report gets built, the data sources, the fallback path, and three caveats)

    line 44
  5. Distinctive:

    Before: (a dense paragraph weighing three options with tradeoffs)

    line 51

“Common” is measured: kept by at least 6 in 10 of the 15 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Plain English

What this is for

Users keep having to tell Claude to redo, rephrase, or "use plain English." The root cause usually isn't formatting. It's that the explanation is too long, buried in jargon, or explaining things they didn't need explained. They bounce off it and have to ask again.

This skill is the fast fix. The user fires it, you rewrite the thing so they actually read it.

What to do

Default target is the last thing you said. If the user points it somewhere else (an email, a doc, text they paste), rewrite that instead.

Rewrite it so:

  • Lead with the answer. First line says the thing or what it means for them. No runway, no "so basically," no "great question."
  • Cut the jargon. If a technical word is genuinely load-bearing, drop one plain phrase next to it in parentheses. Otherwise replace it. Assume the reader is smart but does not want to debug your internals.
  • Shorter than you think. Most of these collapse to 2-4 sentences or a few bullets. If the original was ten lines, the rewrite is probably three.
  • Drop what they can't act on. The caveats, the "it's worth noting," the background they didn't ask for. If it doesn't change what they do next, it goes.
  • Keep it human. No filler, no corporate buzzwords, no "it's not X, it's Y" constructions, short sentences. If it reads like a LinkedIn post, it's wrong.

Then just give the clean version. Don't show the before, don't explain what you trimmed, don't apologize for the first one. They asked for the short version, so the response to "make it shorter" should not itself be long.

The instinct to carry forward

The bigger signal: if the user is invoking this, the last answer missed. Carry that into the rest of the session without being told. Default to shorter and plainer for the rest of the conversation. They can always ask for more depth. They rarely will.

Examples

Example 1, too technical

Before:

The cron routine fires via a scheduled trigger on a weekday cadence and the dedup logic keys off whether the thread already has a draft, so idempotency is preserved across overlapping invocations.

After:

It runs every weekday morning. It won't double-draft the same email because it checks whether a draft already exists first.

Example 2, too long

Before: (eight lines walking through every step of how a report gets built, the data sources, the fallback path, and three caveats)

After:

The report is built from the live numbers, with a screenshot-paste backup if the data source is down. Want me to run one now?

Example 3, they just don't want to read it

Before: (a dense paragraph weighing three options with tradeoffs)

After:

Go with option B. It's the least work and you can change it later. The other two lock you in.

Want the reasoning, or is that enough?

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.