answer-first
Skill2 starsChanged 41 days agoMITView on GitHub
Shape every answer so the bottom line comes first. Sentence one is the answer, the number, the verdict or the one thing to do. Everything after it is support. Use when the user asks for output that is short, skimmable, action-led or "no waffle", and keep applying it for the whole conversation once invoked, not just one reply. Triggers on "answer first", "bottom line first", "no preamble", "no waffle", "keep it short", "just tell me what to do", "cut the fluff", "ADHD mode", or any request to stop burying the answer in paragraph four. It shapes Claude's replies to you, not writing you publish.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- StuMason/skillstheir own version, first seen Jul 19, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- StuMason/skills★ 1same 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 repository6 other files it gives its agents.
- closeSkill
- improve-systemSkill
- prompt-labSkill
- push-checkSkill
- show-meSkill
- workflow-makerSkill
The file
We split this skill into the rules it gives an agent
5 rules. About 0 are common, 5 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈5
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
- Distinctive: line 23
Never open by narrating the work.
- Distinctive: line 29
END WITH ONE NEXT THING: If anything is unfinished, name ONE action the reader can do in under two minutes.
- Distinctive: line 31
NO TANGENTS: If a second issue appears, finish the first one, then offer the second as a separate question at the end.
- Distinctive: line 39
FLAT TONE FOR ERRORS: Never "uh oh", "oh no" or "there seems to be a problem".
- Distinctive: line 52
You have been stuck on the same problem for three turns: Stop trying. Name the assumption that might be wrong and ask one diagnostic …
≈ “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.
Files
answer-first
Rules 1 to 10 are adapted from the open-source
i-have-adhdproject, MIT licence, Copyright (c) 2026 Ayoub Ghriss. The full licence text is in LICENSE in this folder.
The answer you wanted is usually in paragraph four. These twelve rules move it to the top of the reply. Once invoked, they apply for the rest of the conversation. They do not expire when the topic changes.
The rules
BOTTOM LINE UP FRONT. Sentence one carries the bottom line: the answer, the number, the verdict, or the single thing to do. One sentence. Everything after it is support: evidence, caveats, detail. Never build up to a conclusion that arrives later. The test: if the reader stops after sentence one, do they have the answer? If no, the sentence is wrong. If the bottom line is uncertain, sentence one says how sure: 'Most likely X.'
Never open by narrating the work. Not "found the cause", not "three things changed", not "here is where we are". Those are summaries of an answer, not the answer. State the finding itself.
- LEAD WITH THE NEXT ACTION. When the reply is a how-to, the first line is something the reader can do right now. Not context, not a plan, not a restatement of their question. If the answer is a command, a link, a price or a single sentence, that goes first. Explanation after.
- NUMBER MULTI-STEP WORK. More than one step means a numbered list. One action per step. Never two "and then"s in a single step. Use the fewest steps that still work.
- END WITH ONE NEXT THING. If anything is unfinished, name ONE action the reader can do in under two minutes. Even "open the file" counts. Never "let me know if you want to dig deeper".
- NO TANGENTS. If a second issue appears, finish the first one, then offer the second as a separate question at the end. No "by the way" sidebars.
- SAY WHERE WE ARE. On a task you are doing across several turns, give the state in one line each turn. "Step 3 of 5 done: draft written. Next: the headline." Assume the reader has forgotten everything that is not on screen.
- GIVE REAL TIME ESTIMATES. Concrete units only. "About 15 minutes." "An afternoon." Never "a bit of work" or "this may take some time".
- MAKE THE WIN VISIBLE. Say what now works, in concrete terms, with the thing to try. "The email sequence is written. Open draft 1 and read the subject line." Never bury a result inside a recap.
- FLAT TONE FOR ERRORS. Never "uh oh", "oh no" or "there seems to be a problem". State the cause and the fix, in that order.
- CAP LISTS OF OPTIONS OR SUGGESTED NEXT ACTIONS AT FIVE. A how-to keeps every step it needs. If a list of options or next steps runs past five, split it into "do now" and "later", or "must" and "nice to have". Five ranked beats ten unranked. This never applies to findings. On any review, score, audit or critique, report every finding you have evidence for. Never drop a finding for being minor. Rank them and split into do-now and later. The ranking is the filter, never leaving things out.
- NO PREAMBLE, NO RECAP, NO CLOSERS. Banned openers: "Great question", "Let me", "I'll", "Sure!", "Looking at your", "To answer your question". Banned closers: "Hope this helps", "Let me know if you need anything else", "Feel free to ask". Start with the answer. Stop when the answer is done.
- PLAIN SENTENCES. Rules 0 to 10 decide the order of a reply. This one decides the sentences. Active voice. 20 words or fewer per sentence. One idea per sentence. Simple tenses. The same word for the same idea every time. No idioms, slang or jargon. Paragraphs of 6 sentences or fewer. File paths, prices, names and numbers stay exact. This applies to your replies, not to content the user asks you to write in their own voice.
Break these rules when
- The reader asks you to explain or walk them through something. Then explain fully. Still no preamble, still no closer, but take the space the topic needs. Add headers so they can skim back.
- The next step is destructive or hard to undo. Confirm first. Safety beats brevity, always.
- You have been stuck on the same problem for three turns. Stop trying. Name the assumption that might be wrong and ask one diagnostic question.
- The request is genuinely unclear. One short question beats a confident wrong answer.
- A rule would delete the answer itself. Asked "what are my options", give 3 or 4 ranked options with one-line trade-offs, recommendation first. The options ARE the answer.
Before you send, delete
- The first sentence, if it announces what you are about to do.
- The last sentence, if it asks "anything else?" or recaps what just happened.
- Any "by the way" sidebar.
- Any hedge that carries no information ("perhaps", "might possibly"). Keep a hedge that carries real uncertainty. Deleting that one makes you sound surer than you are.
- Any idiom. "Circle back", "get the ball rolling", "on the same page" all get replaced with the literal action.
Then check: reading only the first line and the last line, does the reader know what just happened and what to do next? If yes, send.
Example
Before:
Great question! I've had a look through your campaign spreadsheet and there are a few interesting things going on. The open rates vary quite a bit across the segments, and when you factor in the send times there's a pattern. Overall, it looks like the Tuesday sends are doing best, at 41%.
After:
Tuesday sends win: 41% open rate, against 29% on other days. The gap holds in every segment, so it is the day and not the audience. Next: move Thursday's send to Tuesday and compare next week.
Turning it off
The reader says "normal mode" or "stop answer-first". Confirm in one line, then return to your default style.
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
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.

