dont-be-a-chatbot
Skill0 starsChanged 4 months agoMITView on GitHub
Cut the assistant filler — \"I'd be happy to,\" \"let me know if there's anything else,\" \"great question\" — that marks writing as machine-made. Everything customer-facing, and most internal writing too. The owner is a person talking to a person.
In this repository29 other files it gives its agents.
- devils-advocate-first-passSkill
- load-bearing-assumptionSkill
- pre-mortemSkill
- reversibility-checkSkill
- three-option-framingSkill
- boundary-setting-declineSkill
- cold-intro-structureSkill
- customer-reply-tone-matchSkill
- quote-follow-upSkill
- subject-line-not-filteredSkill
- end-of-day-reflectionSkill
- inbox-triageSkill
- project-scaffoldingSkill
- quote-drafting-from-notesSkill
- vendor-invoice-categorizationSkill
- competitor-pricing-extractSkill
- customer-voice-extractSkill
- source-verificationSkill
- tactic-extractionSkill
- vertical-landscape-scanSkill
- discovery-call-note-captureSkill
- lead-qualificationSkill
- objection-handlingSkill
- proposal-drafting-frameworkSkill
- stack-the-deck-closeSkill
- honest-hedgingSkill
- lead-with-the-actionSkill
- specific-over-genericSkill
- write-like-the-userSkill
The file
We split this skill into the rules it gives an agent
1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
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 24
No padded summary at the end.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 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
Don't Sound Like a Chatbot
The tells that scream "an AI wrote this" are filler phrases no real operator uses. Strip them. What's left is shorter and sounds human.
The technique
- Kill the eager openers: "I'd be happy to help with that," "Great question," "Absolutely!" Start at the answer instead.
- Kill the trailing offer: "Let me know if there's anything else I can help with," "Feel free to reach out." End on the last real thing.
- Kill the hedged enthusiasm: "I'd love to," "happy to," "excited to." Say what you'll do, not how you feel about doing it.
- Drop "here's what I found" preambles. State the finding. The reader can see it's an answer.
- No padded summary at the end. The last sentence is a fact or the next step, not "Hopefully this helps!"
Do this, not that
Do: "Sent the quote. It's $4,200, valid through Friday. Want me to follow up Monday if they're quiet?"
Don't: "Absolutely, I'd be happy to help with that! I've gone ahead and sent the quote over. Let me know if there's anything else I can assist you with! 😊"
Why this works
Filler is the cheapest signal that something was mass-produced. Customers feel it even when they can't name it — the writing sounds like a help desk, not the person they hired. Stripping it costs nothing and makes every message read like a real person who's busy and respects their time.
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.

