agentleFS
Itsme23476Itsme23476/everyday-claude-skills/draft-reply/SKILL.md

draft-reply

Skill1 starsChanged 3 months agoMITView on GitHub

Write a reply to a specific email or message in the user's own voice and save it as a draft — never send. Use when the user says "reply to X", "draft a response to that", "answer this email", or "tell them yes/no" about a specific message.

6 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

5 rules. 0 are common, 5 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 6 edited versions we read keep these.

  • 5

    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:

    Write the reply the user would write, and leave it in drafts for them to send.

    line 8
  2. Distinctive:

    Match the user's voice: If you can see their past sent mail, mirror it: greeting style, length, sign-off, how formal.

    line 15
  3. Distinctive:

    Don't add claims, numbers, commitments, or dates the user didn't give you.

    line 30
  4. Distinctive:

    No filler openers. No "I hope you're well" unless that's genuinely their style.

    line 31
  5. Distinctive:

    If the message is asking for something sensitive (money, credentials, personal data), draft nothing and flag it to the user instead.

    line 32

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

Draft a Reply

Write the reply the user would write, and leave it in drafts for them to send.

Steps

  1. Find the message. Confirm which one if it's ambiguous — sender + subject.
  2. Read the whole thread, not just the last message, so the reply fits the context and doesn't repeat what's already been said.
  3. Match the user's voice. If you can see their past sent mail, mirror it: greeting style, length, sign-off, how formal. When in doubt, go shorter and plainer.
  4. Write the reply. Answer what was actually asked. If the user gave you a direction ("say yes but push the date to Friday"), follow it exactly.
  5. Save it as a draft on the existing thread. Report that it's drafted.

Then

Show the user the draft in full and tell them where it is. Offer one line: "Change anything, or is it good to send?" — but do not send it yourself.

Rules

  • Never send. The user sends. Always.
  • Don't add claims, numbers, commitments, or dates the user didn't give you.
  • No filler openers. No "I hope you're well" unless that's genuinely their style.
  • If the message is asking for something sensitive (money, credentials, personal data), draft nothing and flag it to the user instead.

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.