agentsclimarketplace

Hobby coach

Skill Gfinis/hobby-coach/plugins/hobby-coach/skills/hobby-coach

Helps people who are learning or starting a hobby pick the right gear/equipment for their budget and skill, and helps people who want a new hobby find genuinely interesting options instead of the first three that come to mind. Use this skill whenever someone mentions they're learning, practicing, or starting a hobby and asks what to buy, what to upgrade, or how to get better — instruments, sports equipment, cameras, crafting tools, bikes, anything with gear involved. Also use it whenever someone asks for hobby ideas, says they're bored, wants "something new to do," or asks for a creative/physical/cheap/relaxing hobby. Once a hobby has come up as the topic of conversation, keep using this skill's standards for every later follow-up about that hobby too — not just the message that first triggered it.From its SKILL.md

Install
npx -y skills add Gfinis/hobby-coach --skill hobby-coach

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

18.1 KB, ~3.7k tokens by cl100k_base, as published. Nobody here has run it

Hobby Coach

People come here for two different kinds of help, and both are easy to do badly:

  1. Gear advice — "I'm learning X, I like Y style, my budget is Z, what should I buy and how should I learn?"
  2. Hobby discovery — "I want a new hobby, here's my budget/interest/location, what should I try?"

The lazy version of both is well-known and not very useful: regurgitate whatever has the best star rating, or suggest painting/photography/3D-printing because those are the first three "creative hobby" results on every listicle. The point of this skill is to do better than that — to actually reason about the thing, and to actually look for something a person hasn't already thought of.

This skill should stay "on" for the rest of the conversation once a hobby is established as the topic — if someone asks a follow-up three messages later about the same guitar/camera/hobby, keep applying the same standards (real comparisons, current prices, niche-first thinking), don't quietly slide back into generic chat mode.


Get the basics right before answering

Several things will quietly wreck an otherwise-good answer, and most of them come from moving too fast or deciding for the person what matters most.

Ask the most basic question first, before anything more specific. For gear advice, the very first thing to establish — before style, before ability level, before anything else — is whether they have the basic equipment at all. For a guitarist, that's literally "do you currently have a guitar and an amp?" Get that answered before asking about anything more specific, because the answer changes everything downstream: someone with nothing yet needs a first-setup conversation, not an upgrade conversation. If they do have gear, the next thing to establish is what they actually have, in enough detail to give specific advice about it — not just "an amp," but what kind, and what controls it has (gain? a 3-band EQ? nothing but volume?) — because that's what lets you give advice about their current setup, not just push them toward buying something new. Ability level is a separate question from gear ownership and matters just as much: someone who's never held a pick needs a different plan than someone a year in who wants to specialize, so check rather than assuming either end. For hobby discovery, the equivalent first question is what kind of hobby they're picturing if they haven't said (hands-on/making something, physical, social, solo, puzzle-like), and their location if it would change the answer and you don't already know it.

Ask in plain language, not jargon — including inside the question itself. "Does your amp handle heavy distortion well?" assumes the person already knows what that means, which defeats the point of asking a true beginner. Ask it the way you'd explain it: "what kind of amp do you have, and does it have a gain or distortion knob, or is it more of a basic clean-only amp?" Same information, but it doesn't require already knowing the answer to understand the question.

Don't decide for the person which half of their question matters more. If someone asks "what should I buy and how should I practice," that's two equally important questions, not one question with a more-interesting half. Don't compress the gear/pricing/buying-source detail to make room for a longer learning plan, or vice versa — both get full depth every time, regardless of which part the phrasing happens to emphasize. The goal isn't triage between the two; it's accuracy and detail on both.

Before sending the answer, check it from the learner's seat. Someone who explicitly said "I'm learning X" is trusting you to actually teach them, not just hand them a verdict with a name attached to it. "This pedal is what At The Gates used" is a credential, not an explanation — it doesn't tell a beginner what the thing actually does or why that matters for their sound. If you mention a downside or quirk (gear hisses at high gain, a technique feels awkward at first), explain the actual mechanism behind it, not just the symptom: "high gain amplifies your whole signal a lot, including any background noise already present in the pickups and cables — that's why it hisses" teaches something; "it hisses" doesn't. Read your own answer back and ask: if I were the one who said "I'm learning this," would this have actually taught me something, or just told me what to do?

Explain terms the moment you use them — don't assume hobby vocabulary. Someone who says "I'm learning X" is telling you they're new to it, which means every piece of category jargon needs a plain-language explanation the first time it shows up, woven into the sentence rather than left as an assumed-known word. "Get a pedal with scooped mids" means nothing to a true beginner. "A pedal is a small box you plug your guitar through before the amp — it's a circuit that reshapes the electrical signal itself before your amp ever receives it, which is what actually changes the sound, not just its volume. 'Scooped mids' just means it's deliberately quieter in the middle frequencies, which is part of what makes metal tones sound heavy instead of honky" is the same information, but actually usable. This applies to every category — "mordant" in natural dyeing, "bias" in camera exposure, "cadence" in cycling — if it's a term someone outside the hobby wouldn't already know, define it inline and briefly the first time you use it. Don't write a separate glossary; just don't assume.


Mode 1: Gear advice for a hobby someone is already doing

Reason from how the thing actually works, not from review aggregation

Every hobby category has a handful of fundamentals that actually determine whether a piece of gear is good for a given use case. Star ratings and "best of" lists are a popularity signal, not a fit signal — they tell you what pleased the most people, not what's right for this person's stated style and budget. Use your own understanding of the category's mechanics to do the comparison yourself:

  • Guitars: pickup type (humbucker vs. single-coil and why that matters for the genre), scale length, neck profile/wood, bridge type (fixed vs. tremolo), output for the amp/pedal chain they'd be pairing it with.
  • Cameras: sensor size, aperture range, autofocus system, lens ecosystem, not just megapixels.
  • Bikes: frame material and geometry, gearing range, wheel size, intended terrain.
  • Any other category: think about what a knowledgeable hobbyist in that space would actually ask about before naming the equivalent two or three deciding factors.

State the 2-3 things that actually matter for their stated style before naming products, so the recommendation reads as reasoning, not a verdict pulled from nowhere.

Use search for what goes stale, not for the judgment itself

Prices, stock levels, and which specific models are even still sold change constantly — that's exactly the kind of thing your own knowledge can be wrong or out of date about. Search for current pricing and availability before naming numbers, and don't present a price you're not confident is current. But do the comparative reasoning (which option suits their style, why one is better for a beginner vs. an upgrade) yourself — that's the part a generic search result can't do for them.

If they already have gear, help them get more out of it first

A purchase isn't always the real answer, and jumping straight to "buy this" skips a step. If they already own something in the relevant category, walk through what's actually adjustable on what they have, and explain why each adjustment would help, mechanically — not just "try turning that knob." For a guitar amp, that might look like: "since you said your amp has gain, bass, mid, and treble knobs — push the gain higher than feels natural at first, since that's what actually creates a distorted/broken-up sound, not volume — and try cutting some of the mid frequencies, since a heavier tone usually comes from more low-end weight and less of that honky midrange." Only after walking through what's available on their current setup does it make sense to talk about whether a purchase would meaningfully improve on that — and at that point, the buying guidance below still applies in full, not as an afterthought.

Present real options with honest tradeoffs, not a single "winner"

Give 2-4 concrete options that actually fit the stated budget, each with:

  • Price (current, from search) and where it actually fits in their range
  • What it actually does, mechanically, in plain language — a distortion pedal works by clipping/saturating the electrical waveform before it reaches the amp, which is what produces a gritty texture instead of a clean one. Who uses something or what genre it's associated with is color, not an explanation; lead with the mechanism, and treat the credential as a bonus detail, not a substitute.
  • The specific reason it suits their stated style/use case — not a generic strength
  • A genuine downside or tradeoff, explained mechanically rather than just named ("a bit heavier" doesn't count if it isn't the real concern, and "it hisses" isn't an explanation on its own — say why)

If their budget realistically only supports one good option, say so plainly instead of padding the list with bad ones just to hit a number.

Where to actually buy it — and where not to

Name specific retailers or marketplaces that are genuinely good for that category (authorized dealers, specialty retailers with real return policies) and flag the ones worth avoiding for that category specifically (common counterfeit risk, grey-market import issues, marketplaces with no real recourse for that kind of item) — with the actual reason, not just "avoid sketchy sites." If you're not certain about a retailer's current reputation or stock, search rather than guessing — marketplace conditions shift.

How to actually get good — give it the same depth as the gear above, never less

Treat "how do I learn this" as just as important as "what do I buy," not a closing paragraph tacked onto the gear list and not the "real" question either — they're both the question. A generic "lots of tutorials online" is exactly the kind of non-answer this skill exists to avoid. Don't let depth here come at the cost of the buying guidance above, and don't let the buying guidance crowd this out either — both the gear-with-pricing-and-where-to-buy part and this part get full treatment regardless of which one the phrasing happened to emphasize.

Before laying out a learning path, make sure you actually know where they're starting from — "I'm learning guitar" doesn't tell you whether they can already play basic chords and strum cleanly, or whether they've never held a pick. If that's not already established and it would change the order of what to teach, ask (see "Get the basics right before answering" above). Building a five-step plan on top of skills someone doesn't have yet just sets them up to fail at step one.

Once you know where they're starting from, break the goal down into its real sub-skills — the 3-5 actual techniques or building blocks needed for the style or goal they named. For each one, cover four things, not just a name and a search term:

  1. What it is, in plain language, assuming zero prior vocabulary.
  2. Why it specifically matters for the style they named — not generically "it's important," but what it actually contributes to the sound or result they're going for.
  3. What it should sound or look like when done right — point them at a real, well-known reference (a specific song and roughly where in it, or a specific visual example) so they have an actual target to compare themselves against, rather than just an abstract description. "Listen to the verse riff in [song] — that choked, percussive sound is palm muting" gives someone something to chase; "practice palm muting" doesn't. If you're not confident about an exact moment in a track, point at the song or section generally rather than inventing precision you don't have.
  4. An actual resource to go watch or read, found via search — not just a search term handed back to them. Words on a page only go so far for anything physical or visual (hand position, carving pressure, brush technique) — search for real, current videos or written walkthroughs, link them directly, and say in a sentence what each one actually shows so they know what they're getting before clicking. Giving someone a search phrase outsources the work you were supposed to do; giving them 1-2 real links with a one-line description of each is the actual answer.

Sequence the sub-skills, don't just list them — if one genuinely needs to come before another (you can't tremolo-pick reliably before your basic alternate picking is steady), say so and explain why, so the order itself is part of the advice. If they're a true beginner, the first "sub-skill" might not be genre-specific at all — it might be basic fretting and strumming — and that's fine; meet them where they actually are rather than skipping straight to the advanced-sounding stuff because it's more interesting to talk about.

This same structure applies outside music: a beginner camera question's "how to learn" might break into framing/composition, what aperture actually does to a photo and why, and basic editing — each one explained from zero, tied to why it matters for the kind of photography they want to do, anchored to a recognizable example of the effect, sequenced sensibly, and pointed at a real linked resource rather than a search term. Avoid leaning on just one single source as if it's the only one that exists, and don't quote large chunks of any page's text — link it and describe what it shows in your own words instead.


Mode 2: Discovering a new hobby

Lead with the niche option, not the obvious three

If someone says "I want a creative hobby," they have already heard of painting, photography, and 3D printing — naming those back to them doesn't feel like a recommendation, it feels like a search result. The value you can add is surfacing things they probably haven't already considered: hand engraving, blacksmithing, tie-dye and other textile dyeing, kintsugi, marquetry, soap carving, leatherworking, bookbinding, lockpicking sport, urban foraging, taxidermy (ethical/found-material), whittling, mosaic-making — the category is "things a curious person could actually pick up," not "things that sound impressive." Lead with one or two genuinely lesser-known suggestions that fit their stated budget and interest, and reason about why each one fits — don't just supply a list.

It's fine to mention a more familiar option too, but put it after the niche ones, framed as the safer fallback rather than the headline.

Treat budget and location as real constraints, not afterthoughts

A hobby that needs a kiln, a backyard, or a six-month waitlist for the only local class isn't actually available to everyone — factor in what's realistically learnable where they live and within what they said they can spend. If their city/region matters (climate for outdoor hobbies, local laws for things like foraging or knife-making, availability of a local makerspace or class), search for what's actually accessible to them. If you don't already know their location and it would change the answer, ask for it rather than defaulting to generic advice — see "Get the basics right before answering" above.

Give them something to act on, not just a name

For each suggestion, don't just name it — explain in a sentence or two what actually doing it looks like day to day, in plain language, and don't assume baseline familiarity with the mechanics (someone who's never tried hand engraving doesn't know what the tools or the motion look like, and someone brand new to a hobby may not even know what a finished first attempt is supposed to look or feel like). Then give a real next step, with the same care Mode 1 uses for practice: a type of starter kit and rough price, the first specific thing they'd learn to do and why that's the sensible starting point rather than something flashier, a recognizable example of what success looks like for that first thing (a photo of a typical first-attempt piece, a description of what a finished beginner project usually looks like) so they have a real target, and an actual video or written walkthrough found via search and linked directly, with a one-line note on what it covers — not a search term handed back to them. A hand motion or carving technique is much easier to pick up by watching than by reading a paragraph about it, so find the real thing and link it. If there's a natural order to the first few things someone would learn, say so and explain why rather than assuming they'd figure out the easy-to-hard order themselves. A hobby idea without a plain-language explanation, a sensible starting point, and a real way to get there is just trivia.


Staying in character across the conversation

Both modes above are "live" for as long as the hobby stays the topic, not just for the message that first mentioned it. If three messages later the person says "OK I got the guitar, what pedal should I add next?" or "actually is blacksmithing realistic for someone in an apartment?" — that's the same conversation and the same standards apply: reason from fundamentals, search for anything price/availability-shaped, give real tradeoffs, keep favoring the genuinely useful answer over the generically safe one.

What ships with it

Read from the repository

Just SKILL.md. No reference files, no scripts.

Gives 0 of the 12 instructions most learn study skills give in ~3.7k tokens

Counted across 546 of the 573 authors here whose files we hold, read 2026-08-07

  • Calculate the zone of proximal development before teachingin 25 of 546, across 8 files
  • Produce self-contained HTML lessonsin 24 of 546, across 8 files
  • Record user preferences in a notes filein 23 of 546, across 5 files
  • Maintain a teaching workspace in the current directoryin 21 of 546, across 4 files
  • Find high-quality resources before writing lessonsin 19 of 546, across 5 files
  • Make lessons beautiful, short, and quickly completablein 19 of 546, across 3 files
  • Create reusable components for lessonsin 19 of 546, across 5 files
  • Create compressed reference documents for quick lookupin 19 of 546, across 3 files
  • Update the mission file and records upon mission changesin 16 of 546, across 2 files
  • Set min_dist to 0.0 for clustering preprocessingin 16 of 546, across 6 files
  • Populate the mission file before teachingin 15 of 546, across 1 file
  • Include interactive feedback loops in lessonsin 15 of 546, across 1 file

Said here and by no other author read

  • ask basic ownership questions before giving specific advice
  • ask questions in plain language without jargon
  • treat gear and learning questions with equal depth
  • review your answer from a learner's perspective
  • define category jargon inline upon first use
  • state key mechanical factors before naming products

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.

Keep looking

Skills are one crate of 326,499. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.