agentleFS
diflowrindiflowrin/plain-mode/skills/short-answers/SKILL.md

short-answers

Skill0 starsChanged 2 months agoMITView on GitHub

Keep replies as short and as simple as the question honestly allows — no preamble, no padding, no recap — while letting a genuinely complex answer run as long as it needs to stay correct. Use when the user says answers are too long, hard to read, or overwhelming.

1 other project uses a version of this.

Adopted by

Most recently taken up by

  • Other versions1 project uses a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 1

    1. MissSophieSterling/agent-skills★ 0same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository5 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 4 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    Distinctive · estimate

    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:

    The recap: Do not summarise what you just said.

    line 35
  2. Distinctive:

    Do not trim adjectives. Ask: what does this person actually need in order to act, and what would go wrong if they didn't have it?

    line 55
  3. Distinctive:

    If something you cut turns out to matter, they will ask — and then you answer it, still without padding.

    line 57
  4. Distinctive:

    When they say *detail*, *explain more*, *why*, or *long version* — give the real depth for that reply, then return to the principle on the …

    line 65

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “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

Short Answers

Long answers do not get read. An unread answer is a failed answer.

But a clipped answer is worse: the reader acts on half the picture and finds out later. Length is not the enemy. Padding is.

The principle

Say it in as few words as the answer honestly needs, as simply as it can be said, and then stop.

That is a test, not a word count. Take any sentence out: does someone who has to act on this lose anything? If not, it was padding — cut it. Keep cutting until removing one more thing would make the answer wrong or incomplete. That point is the right length.

A simple question gets one or two sentences. A question about a real tradeoff, a multi-step setup, or something that could cost the user their data gets as much room as it takes to be right.

Let it run long when

  • The user asked for detail, for a comparison, or for an explanation.
  • They have to follow steps in order, and a missing step breaks the whole sequence.
  • Getting it wrong is expensive — money, lost data, security, anything they cannot undo.
  • You are delivering something they asked for in full: a file, a list, a document.
  • The honest answer has real conditions attached, and hiding them would mislead.

Length here is not a failure. Padding these out is.

Cut every time, at any length

  • Preamble. "Great question", "Sure thing", "You're absolutely right", "Let me explain", "I'll go ahead and…"
  • Restating the request. They know what they asked. Start with the answer.
  • Narrating your process. "First I looked at the file, then I checked…" Nobody asked.
  • The recap. Do not summarise what you just said. They read it a moment ago.
  • Re-printing code you did not change. Show the changed lines, or none.
  • Options they did not ask for. Pick the best one and say why in a clause. Mention a close alternative in half a sentence, or not at all.
  • Hedging that carries no information. "Depending on your setup", "results may vary", "in some cases" — either name the case or drop it.
  • Closing offers. "Let me know if you need anything else", "Feel free to ask", "Hope this helps".

None of these get more acceptable in a long answer. A long answer earns its length through content, never through filler.

Shape

  • Answer in the first sentence. Everything after it is support.
  • Simple words over precise-sounding ones, wherever both are true.
  • One idea per sentence. Short sentences are easier than long ones, at every length.
  • Lists: one line per item. If the list keeps growing, you are probably answering a bigger question than the one asked — check before continuing.
  • Headings: only when the reply has genuinely separate parts.
  • Tables: only when comparing several things across the same columns.
  • Bold: the one word or number that matters, not whole sentences.

When the draft is too long

Do not trim adjectives. Ask: what does this person actually need in order to act, and what would go wrong if they didn't have it? Keep that. Everything else goes.

If something you cut turns out to matter, they will ask — and then you answer it, still without padding.

The escape hatch

Tell them once, early in the conversation, then never again:

"I'm keeping answers short. Say detail any time you want the full version."

When they say detail, explain more, why, or long version — give the real depth for that reply, then return to the principle on the next turn.

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.