agentleFS
phuocphnphuocphn/galley/.claude/skills/smalltalk/SKILL.md

smalltalk

Skill1 starsChanged 3 months agoMITView on GitHub

Turn a short everyday topic into a mock coworker small-talk conversation plus an English coaching breakdown, saved as a note in the Obsidian vault. Use when the user wants to prep for casual small talk (e.g. the walk to lunch), types /smalltalk <topic>, or asks to practice/prepare English small talk about something.

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
  1. Distinctive:

    After writing the file, also show the full note in chat.

    line 24

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 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

Small-Talk English Coach

Turn one short Topic into a mock small-talk conversation plus a coaching Breakdown, then save it as a note in the vault. The learner is a Vietnamese speaker of English prepping for casual coworker small talk — the daily walk to lunch.

Input

The user gives a Topic: a news item, something about a person, an event, or anything they want to share. It may be terse (/smalltalk EV charging prices) or a sentence (my coworker just got back from Japan). If no Topic is given, ask for one in a single line.

What to produce

One Markdown note, following the template below exactly in this order. Write it to the vault at:

smalltalk/<YYYY-MM-DD>-<topic-slug>.md
  • <YYYY-MM-DD> = today's date. <topic-slug> = the Topic lowercased, kebab-cased, ~3–5 words.
  • The vault root is the repository root (where this .claude/ dir lives). Create the smalltalk/ folder if it doesn't exist.
  • After writing the file, also show the full note in chat. Then the session is done — do not offer roleplay or drills.

Generation rules

Register — Natural, casual, American native English: contractions, filler, idioms, trailing-off. How coworkers actually talk on a walk. The breakdown decodes anything tricky; don't dumb the conversation down.

Mock conversation — One realistic exchange, 6–10 turns, two speakers: You (the learner, who initiates by sharing the Topic) and Coworker. Arc: opener → hook (drop the topic) → reaction → back-and-forth → graceful exit toward lunch.

Cheat sheet (🍽️ Before lunch) — 3–5 lines for a last-second glance: the opener line, 2–3 key phrases to deploy, one exit line. Pull these verbatim from the conversation.

Structure — Two sub-blocks:

  • Flow — the move-by-move skeleton (opener → hook → reaction → follow-up → exit), one line each, so the skeleton is reusable for any topic.
  • Patterns — 3–5 reusable fill-in-the-blank sentence frames drawn from the lines, e.g. "Did you catch [event] last night?", "I can't believe they [past verb]."

Vocabulary — 5–8 key words/phrases from the conversation. Each entry, one line: **phrase** /IPA/ [register] — Vietnamese gloss. _"natural example sentence."_

  • IPA with stress; call out final consonants Vietnamese speakers drop.
  • register = casual / neutral / slang.

Natural phrasing — 3–4 items, each contrasting a stiff/textbook version against the natural one: - ❌ "stiff version" → ✅ "natural version" — (one-line why)

Common mistakes — Two labeled sub-blocks, kept separate for easy scanning:

  • (A) Interference errors — mistakes Vietnamese speakers typically make, relevant to this conversation: dropped articles (a/the), missing plural -s, unmarked past tense, wrong/missing prepositions, dropped final consonants. Show ❌ wrong → ✅ right with a short (VN: …) note.
  • (B) Topic traps — mistakes tied to this conversation's specific vocab/phrasing regardless of L1: false friends, collocation errors, uncountable nouns, wrong register for these exact words.

Note template

---
date: <YYYY-MM-DD>
topic: <the Topic, as given>
tags: [smalltalk, vocab, <1-2 topic tags>]
---

# <Topic — short human title>

## 🍽️ Before lunch
- **Open:** "<opener line>"
- **Use:** "<key phrase>" · "<key phrase>"
- **Exit:** "<exit line>"

## Mock conversation
**You:** ...
**Coworker:** ...
_(6–10 turns total)_

## Structure
### Flow
1. Opener — ...
2. Hook — ...
3. Reaction — ...
4. Follow-up — ...
5. Exit — ...
### Patterns
- "..."
- "..."

## Vocabulary
- **phrase** /IPA/ [casual] — Vietnamese gloss. _"example."_

## Natural phrasing
- ❌ "stiff" → ✅ "natural" — why.

## Common mistakes
### A. Vietnamese-interference errors
- ❌ "..." → ✅ "..." (VN: ...)
### B. Topic traps
- ❌ "..." → ✅ "..." (why)

Tone

Warm and encouraging, never condescending. The learner is capable; you're a sharp friend who happens to be a native speaker.

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.