agentleFS
EnjoyoerEnjoyoer/loadout/skills/planning-writing/sample/SKILL.md

sample

Skill0 starsChanged 25 days agoMITView on GitHub

Draft or revise WhatsApp, group-chat, and text messages in a style inferred from writing samples the user supplies. Preserve exact facts and ask when a missing fact changes the message.

1 other project uses a version of this.

Adopted by

Most recently taken up by

  • jrswab/axetheir own version, first seen Oct 22, 2026

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:

    Use only facts, recipient context, and approvals supplied in the conversation or verified through an authorized source.

    line 12
  2. Distinctive:

    If a missing fact materially changes the message, ask one short question.

    line 14
  3. Distinctive:

    Do not invent promises, approvals, completed actions, or certainty.

    line 15
  4. Distinctive:

    Use a connector only when the user separately asks to retrieve or verify message facts.

    line 16

≈ “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

Sample-based message voice

Use writing samples supplied in the current conversation to infer the user's tone, sentence length, vocabulary, and level of formality. Do not claim to know a personal voice when no samples are available. If style matching matters and no sample is present, ask for one to three short examples; otherwise use the neutral fallback below.

Facts and authority

  • Use only facts, recipient context, and approvals supplied in the conversation or verified through an authorized source.
  • Preserve exact names, amounts, dates, statuses, and authority boundaries.
  • If a missing fact materially changes the message, ask one short question.
  • Do not invent promises, approvals, completed actions, or certainty.
  • Use a connector only when the user separately asks to retrieve or verify message facts.

Style

Match the user's examples without carrying over typos, private frustration, or profanity unless they are clearly intended for the recipient. For a neutral fallback, start with the status or action, use ordinary words, and keep routine messages to one to three short sentences. A multi-item update may use separate short lines.

Choose the message shape that fits the request:

  • Status: answer first, then add context that changes the next action.
  • Action update: state the attempt, result, and next action.
  • Decision request: name the action and ask a direct question.
  • Correction: state the exact fact and the necessary distinction.
  • Sensitive disagreement: stay calm and narrow; suggest a conversation when the history is disputed.
  • Outside contact: be concise and courteous, with the requested response stated plainly.

Output

Return one sendable version unless the user asks for options. Return only the message text, with no Draft: label, blockquote, analysis, or explanation. Keep separate obligations on separate lines. Before responding, remove any sentence that is more formal than the supplied samples require.

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.