hackathon
Skill2 starsChanged 19 days agoMITView on GitHub
Prepare a winning hackathon submission. Use when the user says "hackathon submission", "submit to hackathon", "demo script", "demo video", "which track should I enter", "Colosseum", "help me win the hackathon", or asks about hackathon grants and Superteam Earn.
4 other projects use a version of this.
Adopted by
Most recently taken up by
- chrislacey89/skillstheir own version, first seen Nov 8, 2026
- anthropics/claude-plugins-officialtheir own version, first seen Nov 8, 2026
- ChainGPT-org/chaingpt-claude-skilltheir own version, first seen Nov 1, 2026
- Dymyt-ry/hackathon-playbooktheir own version, first seen Oct 26, 2026
Other versions4 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 4
- anthropics/claude-plugins-official★ 46ksame name
- ChainGPT-org/chaingpt-claude-skill★ 7same name
- chrislacey89/skills★ 4same name
- Dymyt-ry/hackathon-playbook★ 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 repository11 other files it gives its agents.
- build-statusSkill
- ecosystem-bdSkill
- fundraisingSkill
- idea-sprintSkill
- incident-commsSkill
- launchSkill
- market-sizingSkill
- pitch-deckSkill
- pitch-reviewSkill
- positioningSkill
- pricingSkill
The file
We split this skill into the rules it gives an agent
1 rule. 0 are common, 1 is distinctive, and 0 are very specific to this project.
0
Common
Most of the 4 edited versions we read keep these.
1
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
- Distinctive: line 53
Before scripting the presentation video, read what-separates-winners.md: citing traction and a founder credential separated 43 accelerated …
“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.
Files
<!-- Adapted from sendaifun/solana-new (submit-to-hackathon, apply-grant), MIT © 2026 SendAI and Superteam; full notice in THIRD_PARTY_NOTICES.md. Telemetry removed. -->Hackathon Submission
Track choice → scannable description → two scripts (2–3-min pitch, ≤3-min demo) → checklist. Optimize for a judge who has 90 seconds, not a reader who has 10 minutes.
Detail lives in references/. The remaining upstream links (colosseum-copilot, the Superteam ideas dataset) are in upstream-packs.md.
Context handoff
At start, read .claude/context/idea.md and .claude/context/build.md (format, written by build-status) if present — pull the pitch, wedge, and what actually works (its "What works today" section) from them instead of asking again. Read .claude/context/positioning.md (format, written by positioning) if present too — the tagline is its plain one-liner, the alternatives table feeds "the novel part", and its words-to-avoid list applies to the description.
Workflow
1. Check the format, then pick the least-crowded track
Colosseum runs its hackathons as startup competitions, and formats change: Frontier 2026 had no tracks at all. Read the current rules first. Where tracks exist, winning a thin track beats placing in a fat one. Per candidate track: estimate entry volume, fit with what's actually built, and judge appetite (sponsor tracks often have the fewest serious entries).
- Winner history: winners.md — Grand Champions and track first places from Grizzlython 2023 to Frontier 2026, dated, each row linked to its announcement post
- Live crowdedness check: colosseum-copilot — query 5,400+ past submissions for cluster density and gaps. CLI-backed: sign in once per machine with
npx @colosseum-org/copilot-connect login(Node 20+).
2. Write a scannable description
Judges read 100+ submissions. Yours gets one skim deciding whether it gets a real read:
- Tagline: what it does, one sentence, no jargon
- First paragraph: problem + who has it
- Bold the one thing that's novel
- "What works today" list — demo-able claims only, never roadmap dressed as product
- Why Solana (one concrete reason: speed, fees, composability with X)
Full structure (200–500 words, paragraph-by-paragraph) and the fields Colosseum's portal asks for, including go-to-market and demand validation: submission-guide.md. Score the draft against judging-criteria.md before submitting: Colosseum's seven published factors, which carry no published weights.
3. Videos (each ≤ 3 minutes)
Colosseum asks for two: a 2–3 minute presentation video (the startup pitch, judged first) and a product demo of at most 3 minutes. Scripts for both: demo-video-script.md. The product demo, in outline:
| Time | Beat |
| 0:00–0:15 | What you're about to show, one sentence |
| 0:15–2:00 | One happy path in the real product, real data, on-chain proof (explorer tx) |
| 2:00–2:40 | One technical highlight and why you built it that way (Solana integration, on-chain logic) |
| 2:40–3:00 | What works today vs. next; repo and live link on screen |
Rule: if the demo can fail live, record it.
Before scripting the presentation video, read what-separates-winners.md: citing traction and a founder credential separated 43 accelerated winners from 47 projects from the same hackathons that won nothing; for the product on screen and an explorer transaction the study detected no difference. The script checklist is pitch-rhetoric.md.
4. Submission checklist
- Track (if the event has tracks) chosen by fit, then crowdedness, not vanity
- Tagline passes the "non-crypto friend" test
- Description scannable in 90 seconds (bold claims, short paragraphs)
- Presentation video 2–3 min (team, problem, evidence, why now)
- Demo video ≤ 3 min, real transaction shown
- Go-to-market and demand-validation fields filled with sourced evidence
- Pre-existing code disclosed; every link opens for a logged-out judge
- Repo public (or access granted to judges), README quickstart actually works from clone
- Deployed link (devnet OK) + program ID listed
- Team and contact info complete
- Pitch deck attached if track requires one — use pitch-deck
After the hackathon: grants
Losing the track doesn't mean losing the funding. Same artifacts (description, demo, deck) feed grant applications:
- Superteam Earn (earn.superteam.fun) — bounties + grants up to ~$10k USDC equivalent, fast cycles, regional Superteams
- Solana Foundation grants — milestone-based, public-good angle; reuse the scannable description with an ecosystem-benefit paragraph
- Superteam Agentic Engineering grant — 200 USDG toward an AI coding subscription, half on approval and half once a live Solana MVP ships (page; applications showed as paused on 2026-10-05)
- Grant-shaped ideas dataset: superteam-ideas.json
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
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.