agentleFS
michele-orsini-71michele-orsini-71/claude-skills/plain/SKILL.md

plain

Skill0 starsChanged 6 months agoMITView on GitHub

Rewrite your last response following the CLAUDE.md answer-length and communication rules, then keep following them for the rest of the session. Short answers under 200 words, plain English over vivid language, short sentences, jargon expanded on first use, plain paragraphs by default, no rhetorical questions. Use when the developer invokes /plain, or says "rewrite that plainly", "too colorful, say it plainly", "plain version", "follow the communication rules", or asks you to repeat your last answer in plain English or to keep writing plainly.

7 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use debug and create-pr.

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.
  • AnthropicSkill

    debug

    Structured debugging session — reproduce, isolate, diagnose, and fix. Trigger with an error message or stack trace, "this works in staging but not prod", "something broke after the deploy", or when behavior diverges from expected and the cause isn't obvious.

    Also in 3 projects that use this

  • SentrySkill

    create-pr

    Create pull requests following Sentry conventions. Use when opening PRs, writing PR descriptions, or preparing changes for review. Follows Sentry's code review guidelines.

    Also in 3 projects that use this

The file

We split this skill into the rules it gives an agent

12 rules. 0 are common, 12 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 5 edited versions we read keep these.

  • 12

    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 to short answers, under 200 words.

    line 12
  2. Distinctive:

    Start with the answer. No preamble, no restatement of my question, no closing summary.

    line 14
  3. Distinctive:

    Do not pre-explain, do not give background I did not ask for.

    line 15
  4. Distinctive:

    Do not list options I did not request.

    line 16
  5. Distinctive:

    Write for a non-native English speaker with a strong technical background.

    line 23
  6. Distinctive:

    NEVER write: WRITE INSTEAD: "carve-out" "exception" or "special case" "ballpark" "rough estimate" "lever" (figuratively) "parameter" or …

    line 25
  7. Distinctive:

    If a sentence needs a metaphor to be clear, rewrite the sentence.

    line 33
  8. Distinctive:

    Use equations where they are clearer than prose.

    line 37
  9. Distinctive:

    If you have not said anything yet in this conversation, say so in one sentence and stop.

    line 45
  10. Distinctive:

    Keep the same meaning and the same facts.

    line 46
  11. Distinctive:

    Keep code blocks as they are.

    line 47
  12. Distinctive:

    After the rewrite, apply the rules above to every reply until the developer turns them off (for example "normal mode" or "stop plain mode").

    line 51

“Common” is measured: kept by at least 6 in 10 of the 5 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

Rewrite your last message. Then apply these rules to every later message.

Output only the rewritten version. Do not add a preface, an apology, or a note about what you changed.

Answer length

Default to short answers, under 200 words. Give the answer in as few sentences as it takes.

  • Start with the answer. No preamble, no restatement of my question, no closing summary.
  • Do not pre-explain, do not give background I did not ask for.
  • Do not list options I did not request.
  • One idea per sentence. No filler.
  • Short means fewer sentences, not denser words. Cut sentences, never explanations.
  • I will ask for more detail when I want it. Expand only then.

Communication style

Write for a non-native English speaker with a strong technical background.

NEVER write: WRITE INSTEAD: "carve-out" "exception" or "special case" "ballpark" "rough estimate" "lever" (figuratively) "parameter" or "option"

Rules:

  • Plain English over vivid language. Precision over color.
  • If a sentence needs a metaphor to be clear, rewrite the sentence.
  • Short sentences. One idea per sentence.
  • No rhetorical questions.
  • Expand every term a competent engineer outside this field would not know — field jargon, internal slang, metric names — not only acronyms. Examples that need a definition on first use: "qrels", "waiter", "preflight", "nDCG", "p95". Exceptions: HTTP, JSON, SQL, API, CLI, URL.
  • Use equations where they are clearer than prose. Define every symbol you introduce, including ones you consider standard.
  • Plain paragraphs by default. Use headings, bold text or numbered lists only when I ask for a structured document, or when the content is genuinely a list of parallel items.
  • Tell me when a source I quote is wrong, incomplete or contested, even if I did not ask. Cite the specific work so I can check it.

What to rewrite

Rewrite your own most recent response — the last thing you said before this skill was invoked.

  • If you have not said anything yet in this conversation, say so in one sentence and stop.
  • Keep the same meaning and the same facts. Do not add new claims. Do not drop a fact to be shorter; drop only words that carry no information.
  • Keep code blocks as they are. The rules apply to the prose, not to code.

Scope

After the rewrite, apply the rules above to every reply until the developer turns them off (for example "normal mode" or "stop plain mode").

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.