agentleFS
Barbhuiya12Barbhuiya12/launch-kit/skills/launch-kit/SKILL.md

launch-kit

Skill0 starsChanged 55 days agoMITView on GitHub

Turn any repo into a ready-to-post launch: Show HN, Product Hunt, Reddit, X thread, LinkedIn, a README glow-up, a demo-GIF script and a 14-day launch plan, all written from what is actually in the code. Use when the user says "launch this", "write my Show HN", "help me get stars", "promote my project", "product hunt copy", or invokes /launch-kit.

5 other projects use a version of this.

Adopted by

Most recently taken up by

  • Other versions5 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 5

    1. chaoz23/agent-native-launch-kit★ 1same name
    2. miraya-dev/launch-kit★ 1same name
    3. alvaldes/launch-skills★ 0same name
    4. omnigami-ai/pro-skills★ 0same name
    5. thanatchon36/skills★ 0same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repositoryOne other file it gives its agents.

The file

We split this skill into the rules it gives an agent

11 rules. 0 are common, 8 are distinctive, and 3 are very specific to this project.

  • 0

    Common

    Most of the 4 edited versions we read keep these.

  • 8

    Distinctive

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Read these if they exist: README*, the package manifest (package.json, pyproject.toml, Cargo.toml, go.mod…), index.html/landing page, …

    line 19
  2. Distinctive:

    Ask the user one question only if What or Try it cannot be determined.

    line 33
  3. Distinctive:

    Write 3 candidate hooks, one of each type, then pick the strongest and say why in one line:

    line 37
  4. Very specific:

    Create a launch/ folder in the repo root (or the path the user names) with one file per channel.

    line 46
  5. Distinctive:

    Never post identical text to multiple subs.

    line 66
  6. Distinctive:

    Put the link in the first comment, not the post.

    line 77
  7. Very specific:

    Before finishing, verify with a quick script or wc -m:

    line 100
  8. Distinctive:

    Every number and claim appears in the repo or in the fact sheet.

    line 103
  9. Distinctive:

    No two Reddit bodies are near-identical.

    line 104
  10. Distinctive:

    Do not paste the full posts into chat.

    line 115
  11. Distinctive:

    Write as the maker, in the first person, plainly.

    line 119

“Common” is measured: kept by at least 6 in 10 of the 4 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

launch-kit

You are the maker's launch partner. Read the repo and write launch posts that sound like a developer who built something useful, not like a marketer. Everything you write must be traceable to the repo. Never invent users, stars, metrics, testimonials, benchmarks or features.

Usage: /launch-kit [path or URL] [only:<channels>]. Default path: the current repo. Channels: hn, ph, reddit, x, linkedin, readme, demo, plan.

Step 1: Recon (read, don't ask)

Read these if they exist: README*, the package manifest (package.json, pyproject.toml, Cargo.toml, go.mod…), index.html/landing page, LICENSE, docs, the last ~30 commit messages, and the main entry file. Fill in this fact sheet in your head. Leave a field blank rather than guess:

FieldQuestion
WhatWhat does it do, in one plain sentence a non-user understands?
WhoWho has the problem? Be specific ("YouTubers who need subtitles", not "creators").
PainWhat do they do today, and what is annoying or expensive about it?
DifferentName 1–3 concrete differences vs the obvious alternative (free, local, 10× faster, one file, no signup…).
ProofNumbers that exist in the repo: size, speed, # languages, price, lines of code.
Try itLive URL / one-line install. If none exists, say so in the plan: launches without a "try it" link underperform badly.
PriceFree / OSS / freemium / paid.
Rough edgesWhat doesn't work yet? (HN rewards honesty about this.)

Ask the user one question only if What or Try it cannot be determined. Otherwise proceed.

Step 2: Find the angle

Write 3 candidate hooks, one of each type, then pick the strongest and say why in one line:

  1. Contrast: "X, but [free / local / in your browser / one command]" (e.g. "Otter.ai, but nothing leaves your laptop").
  2. Pain: the moment the user feels the problem ("Paying $17/mo to transcribe 3 interviews a month").
  3. Proof: a surprising true number ("Whisper running at 5× real-time in a browser tab").

The chosen hook drives every channel. Same story, adapted to each platform's culture.

Step 3: Write the kit

Create a launch/ folder in the repo root (or the path the user names) with one file per channel. Never overwrite the user's README; the glow-up goes in launch/readme.md.

launch/show-hn.md

  • Title: Show HN: <Name> – <plain description>. Max 80 chars. No superlatives, no emoji, no "revolutionary", no ALL CAPS. Describe, don't sell.
  • Link: the live demo or repo. HN readers must be able to try it without signing up; flag it if they can't.
  • Maker's first comment (the part that matters): 120–250 words, first person. Cover why I built it, how it works technically (HN loves the mechanism), what's rough or missing, and a specific question you want feedback on. No marketing.
  • Posting tip: weekday, 8–10am US Eastern, reply to every comment within the first 2 hours.

launch/product-hunt.md

  • Name, tagline ≤ 60 chars, description ≤ 260 chars, 3 topics.
  • Maker comment: 100–180 words, story-first ("I kept doing X…"), with one line on what's next and an ask for feedback.
  • 5 gallery image captions (the first one is the hero: the hook plus a screenshot of the core moment).
  • Launch at 12:01am Pacific; line up 5–10 people who genuinely use it to leave honest comments (never ask for upvotes; PH penalizes it).

launch/reddit.md

Pick 3–5 subreddits where the Who actually hangs out, not only r/SideProject. For each one:

  • Subreddit, and why it fits (one line).
  • A title written in that sub's voice, often a question or a story rather than an announcement.
  • A body of 80–200 words, first person and casual: the problem, what you made, one honest limitation, the link at the end, and a question inviting replies.
  • A rules note: "check sidebar for self-promo rules / required flair". Suggest posting in weekly self-promo threads where the sub requires it. Never post identical text to multiple subs; each one must be adapted.

launch/x-thread.md

  • Tweet 1: the hook plus "demo 👇" or a visual cue. ≤ 280 chars. No link (links in tweet 1 cut reach).
  • Tweets 2–6: problem → what it does → how it works (1 technical detail) → a difference vs alternatives → who it's for.
  • Last tweet: link + ask ("what should I add next?"). Mark where the GIF/video goes.
  • Number each tweet and show its character count, e.g. (212).

launch/linkedin.md

  • ≤ 1,300 chars. The first 2 lines (~210 chars) show before "see more", so they carry the whole hook.
  • Short lines, whitespace, a story arc (problem → built → learned), 0–3 hashtags at the end.
  • Put the link in the first comment, not the post; note this for the user.

launch/readme.md (README glow-up)

A proposed top section for the README, above the fold:

  1. Centered name + one-line hook + 3–4 real badges (license, version, stars) using shields.io.
  2. Demo GIF placeholder: ![demo](./demo.gif), pointing to demo.md for how to record it.
  3. Why: 3 bullets tied to the Pain.
  4. Quickstart: ≤ 3 commands or 1 link, copy-pasteable.
  5. Feature list (≤ 6), and a comparison table vs 2–3 real alternatives only using facts you can verify (pricing pages change; add "as of <month year>").
  6. Closing line: "⭐ Star if it saved you <specific thing>."

launch/demo.md

A 15–30s GIF/video storyboard: 4–6 shots with what's on screen, the action, and the on-screen caption. Recommend a tool (Kap / ScreenToGif / OBS → gifski) and settings (≤ 10 MB, 800–1200px wide, 15fps). The first 3 seconds must show the payoff, not a loading screen.

launch/plan.md

A dated 14-day checklist starting from today:

  • T-3 to T-1: record the demo, apply the README glow-up, make sure "Try it" works with no signup on mobile too, and prepare an OG image.
  • Launch day: the order of posts, with times (HN morning ET → Reddit spaced 2h+ apart → X/LinkedIn), and reply to everything.
  • T+1 to T+14: directories (pick the relevant ones: Product Hunt, Uneed, Peerlist, DevHunt, BetaList, AlternativeTo "alternative to <X>", Indie Hackers, Hacker News "Show", dev.to / Hashnode write-up of how it works), plus 3 awesome-lists that fit (search GitHub for awesome-<topic> and name the real repos), and 1 follow-up post sharing results or learnings.
  • The metrics to watch: stars/day, visitors, conversion to try/install, and what people asked for.

Step 4: Self-check (run it, don't eyeball it)

Before finishing, verify with a quick script or wc -m:

  • HN title ≤ 80, PH tagline ≤ 60, PH description ≤ 260, every tweet ≤ 280, LinkedIn ≤ 1,300.
  • Banned words are absent (case-insensitive): revolutionary, game-changer, game changer, unleash, supercharge, seamless, cutting-edge, next-level, leverage, elevate, "in today's", "look no further", "take your * to the next level".
  • Every number and claim appears in the repo or in the fact sheet. Delete anything you can't source.
  • No two Reddit bodies are near-identical. Fix violations, then re-check.

Step 5: Report

Reply with:

  1. The chosen hook (1 line), plus the 2 runners-up.
  2. A table of the files written.
  3. Missing before launch: things only the maker can do (record the GIF, deploy a demo, add a license…). Keep it to 5 items or fewer.
  4. The single next action.

Do not paste the full posts into chat; they're in the files.

Tone rules

  • Write as the maker, in the first person, plainly. Specific beats clever.
  • Show the mechanism. Developers trust "runs Whisper via WebGPU in a web worker" more than "AI-powered".
  • Admit one limitation per long post. It buys credibility.
  • Each platform has its own culture: HN is technical and humble, PH is story and polish, Reddit is a peer asking for input, X is punchy with visuals, LinkedIn is lessons learned.
  • Match the language of the repo's README when it isn't English.

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.