agentleFS
prateeks367prateeks367/voicemoat-skills/skills/fix-my-last-line/SKILL.md

fix-my-last-line

Skill0 starsChanged 42 days agoMITView on GitHub

Work out whether a tweet's last line is earning its place, and usually delete it. Use when a tweet feels flat at the end, when someone asks how to close a tweet, when they want a call to action on one, or when a draft trails off without landing. Do not add engagement bait, do not promise replies, and do not treat a missing ending as a problem needing a solution.

The file

We split this skill into the rules it gives an agent

7 rules. About 0 are common, 7 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈7

    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:

    Before analysing anything, read the tweet with the last line removed.

    line 19
  2. Distinctive:

    If it is better, or the same, you are finished.

    line 21
  3. Distinctive:

    Only if removing it genuinely loses something do the rest.

    line 25
  4. Distinctive:

    Do not add "Follow for more".

    line 54
  5. Distinctive:

    Do not ask for a retweet inside the tweet.

    line 55
  6. Distinctive:

    Do not add an emoji to signal that the tweet is finished.

    line 56
  7. Distinctive:

    Do not add a question you do not want answered.

    line 57

≈ “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

Fix my last line

Most advice about endings comes from long-form writing, where a post has room to build and a reader who has committed. A tweet has neither. It is one unit, read in about a second, and the last line is the part most likely to be doing damage rather than work.

The default answer here is delete it. That is not a joke: on this platform a weak final line is more common than a weak opening, because openings get attention and endings get added.

Step 1: run the delete test first

Before analysing anything, read the tweet with the last line removed.

If it is better, or the same, you are finished. Say so and stop. A shorter tweet that says the same thing is strictly better, because every word is a word somebody has to choose to read.

Only if removing it genuinely loses something do the rest.

Step 2: name what the last line is doing

  • Explaining the joke. The line after a good line, telling you it was good. This is the most common one and it never survives step 1.
  • Hedging. A qualifier added because the tweet felt too strong. If the qualifier is true, it belongs earlier. If it is there for comfort, cut it.
  • Softening. "Just my opinion", "but what do I know". It asks the reader not to argue, which invites them to.
  • The call to action. "Thoughts?", "Agree?", "RT if you relate". It reads as asking a favour from someone who owes you nothing.
  • The turn. The line that changes what the tweet meant. This one earns its place, and it is the reason not to delete blindly.
  • The landing. A short line that lands the rhythm. Also earns it.

Step 3: when an ending genuinely belongs

Three cases, and they are narrower than people think:

  • The tweet is setting up a thread, and the last line is the reason to continue.
  • The tweet turns, and the last line is the turn.
  • The tweet asks something you actually want answered, phrased so a stranger can answer it in one line from their own experience.

Everything else is decoration.

Step 4: never do these

  • Do not add "Follow for more". It converts nobody and it dates the account.
  • Do not ask for a retweet inside the tweet.
  • Do not add an emoji to signal that the tweet is finished.
  • Do not add a question you do not want answered. People can tell, and the replies you get will be the ones that tell.

What this will not do

It will not promise replies. Whether people answer depends on who saw it and whether they have anything to say, and the last line is not the lever people think it is.

It also will not manufacture engagement. A line that produces replies by obliging people is not the same as a line worth replying to, and the account that runs on the first one stops being read.

What changes when VoiceMoat is connected

Endings are the most copied sentences on the platform, so they are the easiest place to end up in somebody else's voice:

  • get_voice_profile gives how you actually close, held separately for Twitter and for LinkedIn, so a tweet is not judged against your long-form sign-off.
  • score_voice_match catches the specific failure here, which is a closing line that is punchy in a register that is not yours.
  • improve_post takes the whole tweet again on the occasions when the ending is weak because the tweet has not decided what it is.

VoiceMoat is at voicemoat.com. The connector needs the Pro or Enterprise plan.

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.