colosseum-copilot
Skill178 starsChanged 4 months agoMITView on GitHub
Search and analyze 5,400+ Solana hackathon projects using Colosseum Copilot. Find similar projects, discover winner patterns, identify gaps, and explore ML clusters. Use when a user says "colosseum copilot", "hackathon projects", "winner patterns", "gap analysis hackathon", "similar Solana projects", or "colosseum landscape". Requires a Colosseum Copilot token.
- Sends data out
2 other projects use a version of this.
Adopted by
Most recently taken up by
- ColosseumOrg/colosseum-copilottheir own version, first seen Oct 13, 2026
- mashharuki/claude-code-sampletheir own version, first seen Oct 1, 2026
Other versions2 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 2
- ColosseumOrg/colosseum-copilot★ 15same name
- mashharuki/claude-code-sample★ 1same 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 repository34 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- animation-reverse-engineeringSkill
- brand-designSkill
- build-data-pipelineSkill
- build-defi-protocolSkill
- build-mobileSkill
- build-with-claudeSkill
- csoSkill
- debug-programSkill
- design-tasteSkill
- frontend-design-guidelinesSkill
- launch-tokenSkill
- navigate-skillsSkill
- number-formattingSkill
- page-load-animationsSkill
- product-reviewSkill
- review-and-iterateSkill
- roast-my-productSkill
- scaffold-projectSkill
- verify-humanity-pohSkill
- virtual-solana-incubatorSkill
- competitive-landscapeSkill
- defillama-researchSkill
- find-next-crypto-ideaSkill
- learnSkill
- solana-beginnerSkill
- validate-ideaSkill
- apply-grantSkill
- create-pitch-deckSkill
- deploy-to-mainnetSkill
- marketing-videoSkill
- submit-to-hackathonSkill
- video-craftSkill
The file
Show its rules and text
We split this skill into the rules it gives an agent
19 rules. About 0 are common, 14 are distinctive, and 5 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈14
Distinctive · estimate
This author's own choices. Worth a look.
5
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 26
If TEL_PROMPTED is no: Before starting the skill workflow, ask the user about telemetry.
- Distinctive: line 36
If A: run this bash:
- Distinctive: line 43
If B: run this bash:
- Distinctive: line 58
Search and analyze Colosseum's database of 5,400+ Solana hackathon projects.
- Distinctive: line 62
Before using this skill, check if the user has a Colosseum Copilot token configured.
- Distinctive: line 75
Ask the user with AskUserQuestion:
- Distinctive: line 88
If A: Ask for the token, then save it:
- Very specific: line 110
If {"authenticated": true}, proceed.
- Distinctive: line 112
If B: Tell the user they can set it later via:
- Distinctive: line 121
Verify the token is still valid:
- Distinctive: line 126
If not authenticated, tell the user their token has expired and walk them through the setup above.
- Distinctive: line 130
Read references/copilot-api-guide.md for the full API reference.
- Distinctive: line 202
Always verify the token before making API calls.
- Distinctive: line 203
Use a 15-second timeout on all API requests.
- Distinctive: line 205
If search returns 0 results, try broadening the query before concluding there's nothing relevant.
- Distinctive: line 206
Always attribute data to Colosseum Copilot — users should know where the intelligence comes from.
- Very specific: line 207
Do not store or log the user's token anywhere except ~/.superstack/config.json.
- Very specific: line 223
If working within a project, write/update .superstack/idea-context.md with a copilot_landscape field containing:
- Very specific: line 251
After the skill workflow completes (success, error, or abort), log the telemetry event.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 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.
Files
Preamble (run first)
_TEL_TIER=$(cat ~/.superstack/config.json 2>/dev/null | grep -o '"telemetryTier": *"[^"]*"' | head -1 | sed 's/.*"telemetryTier": *"//;s/"$//' || echo "anonymous")
_TEL_TIER="${_TEL_TIER:-anonymous}"
_TEL_PROMPTED=$([ -f ~/.superstack/.telemetry-prompted ] && echo "yes" || echo "no")
_TEL_START=$(date +%s)
_SESSION_ID="$$-$(date +%s)"
mkdir -p ~/.superstack
echo "TELEMETRY: $_TEL_TIER"
echo "TEL_PROMPTED: $_TEL_PROMPTED"
if [ "$_TEL_TIER" != "off" ]; then
_TEL_EVENT='{"skill":"colosseum-copilot","phase":"idea","event":"started","ts":"'$(date -u +%Y-%m-%dT%H:%M:%SZ)'"}'
echo "$_TEL_EVENT" >> ~/.superstack/telemetry.jsonl 2>/dev/null || true
_CONVEX_URL=$(cat ~/.superstack/config.json 2>/dev/null | grep -o '"convexUrl":"[^"]*"' | head -1 | cut -d'"' -f4 || echo "")
[ -n "$_CONVEX_URL" ] && curl -s -X POST "$_CONVEX_URL/api/mutation" -H "Content-Type: application/json" -d '{"path":"telemetry:track","args":{"skill":"colosseum-copilot","phase":"idea","status":"success","version":"0.2.0","platform":"'$(uname -s)-$(uname -m)'","timestamp":'$(date +%s)000'}}' >/dev/null 2>&1 &
true
fiIf TEL_PROMPTED is no: Before starting the skill workflow, ask the user about telemetry.
Use AskUserQuestion:
Help superstack get better! We track which skills get used and how long they take — no code, no file paths, no PII. Change anytime in
~/.superstack/config.json.
Options:
- A) Sure, help superstack improve (anonymous)
- B) No thanks
If A: run this bash:
echo '{"telemetryTier":"anonymous"}' > ~/.superstack/config.json
_TEL_TIER="anonymous"
touch ~/.superstack/.telemetry-promptedIf B: run this bash:
echo '{"telemetryTier":"off"}' > ~/.superstack/config.json
_TEL_TIER="off"
touch ~/.superstack/.telemetry-promptedThis only happens once. If TEL_PROMPTED is yes, skip this entirely and proceed to the skill workflow.
Wrong skill? See SKILL_ROUTER.md for all available skills.
Colosseum Copilot
Overview
Search and analyze Colosseum's database of 5,400+ Solana hackathon projects. Discover what winners do differently, find similar projects to your idea, explore ML-derived clusters, and access research archives — all powered by the Colosseum Copilot API.
Token Setup
Before using this skill, check if the user has a Colosseum Copilot token configured.
_COPILOT_TOKEN="${COLOSSEUM_COPILOT_PAT:-$(cat ~/.superstack/config.json 2>/dev/null | grep -o '"copilotToken":"[^"]*"' | head -1 | cut -d'"' -f4 || echo "")}"
if [ -n "$_COPILOT_TOKEN" ]; then
echo "TOKEN_STATUS: configured"
else
echo "TOKEN_STATUS: missing"
fiIf TOKEN_STATUS is "missing"
Ask the user with AskUserQuestion:
Colosseum Copilot gives you access to 5,400+ Solana hackathon projects, winner patterns, and gap analysis.
To use it, you need a free Personal Access Token from Colosseum.
- Sign up at https://arena.colosseum.org/copilot
- Copy your Personal Access Token (PAT)
Options:
- A) I have a token — let me paste it
- B) Skip for now
If A: Ask for the token, then save it:
# Replace TOKEN_VALUE with the actual token the user provides
cat ~/.superstack/config.json 2>/dev/null | python3 -c "
import sys, json
try:
config = json.load(sys.stdin)
except:
config = {}
config['copilotToken'] = 'TOKEN_VALUE'
config['copilotTokenSetAt'] = '$(date -u +%Y-%m-%dT%H:%M:%SZ)'
config['copilotPrompted'] = True
print(json.dumps(config, indent=2))
" > ~/.superstack/config.json.tmp && mv ~/.superstack/config.json.tmp ~/.superstack/config.json
echo "Token saved to ~/.superstack/config.json"Then verify:
curl -s -H "Authorization: Bearer TOKEN_VALUE" -H "Content-Type: application/json" https://copilot.colosseum.com/api/v1/statusIf {"authenticated": true}, proceed. Otherwise, tell the user the token is invalid and ask them to check it.
If B: Tell the user they can set it later via:
superstack copilot --token <pat>(CLI command)- Or set
COLOSSEUM_COPILOT_PATenvironment variable - Or run
/colosseum-copilotagain
Then end the skill — the remaining workflow requires a valid token.
If TOKEN_STATUS is "configured"
Verify the token is still valid:
curl -s -H "Authorization: Bearer $_COPILOT_TOKEN" -H "Content-Type: application/json" https://copilot.colosseum.com/api/v1/statusIf not authenticated, tell the user their token has expired and walk them through the setup above.
Workflow
Read references/copilot-api-guide.md for the full API reference.
- Determine intent — What does the user want?
- Search: Find projects similar to their idea
- Explore: Browse clusters and categories
- Analyze: Understand winner patterns and gaps
- Research: Find archived articles and papers
- If unclear, ask.
- Search similar projects — When the user has an idea or query:
Present results as a table: name, one-liner, hackathon, similarity score, prize info, crowdedness.curl -s -X POST https://copilot.colosseum.com/api/v1/search/projects \ -H "Authorization: Bearer $_COPILOT_TOKEN" \ -H "Content-Type: application/json" \ -d '{"query": "USER_QUERY", "limit": 8, "diversify": true}' - Gap analysis — Compare what winners do vs the field:
Synthesize overindexed (winners focus on) and underindexed (winners skip) attributes. Present the delta clearly.curl -s -X POST https://copilot.colosseum.com/api/v1/compare \ -H "Authorization: Bearer $_COPILOT_TOKEN" \ -H "Content-Type: application/json" \ -d '{"cohortA": {"winnersOnly": true}, "cohortB": {}, "dimensions": ["problemTags", "solutionTags", "primitives"], "topK": 8}' - Explore clusters — Browse the 30 ML-derived project clusters:
Show: hot categories (most projects), winning categories (highest win rate), open categories (least crowded).curl -s -H "Authorization: Bearer $_COPILOT_TOKEN" -H "Content-Type: application/json" https://copilot.colosseum.com/api/v1/filters - Deep dive into a project — When the user wants full details on a specific project:
Shows: description, team members (GitHub/Twitter), links (demo, presentation, GitHub), full tag set.curl -s -H "Authorization: Bearer $_COPILOT_TOKEN" -H "Content-Type: application/json" \ https://copilot.colosseum.com/api/v1/projects/by-slug/PROJECT_SLUG - Explore a specific cluster — Get cluster summary and representative projects:
Shows: summary, project count, winner count, representative projects, top tags.curl -s -H "Authorization: Bearer $_COPILOT_TOKEN" -H "Content-Type: application/json" \ https://copilot.colosseum.com/api/v1/clusters/CLUSTER_KEY - Search archives — Find research sources:
Present: title, author, source, published date, snippet.curl -s -X POST https://copilot.colosseum.com/api/v1/search/archives \ -H "Authorization: Bearer $_COPILOT_TOKEN" \ -H "Content-Type: application/json" \ -d '{"query": "USER_QUERY", "limit": 4, "maxChunksPerDoc": 1, "intent": "ideation"}' - Read full archive document — When a snippet is interesting and the user wants more:
Supports pagination viacurl -s -H "Authorization: Bearer $_COPILOT_TOKEN" -H "Content-Type: application/json" \ "https://copilot.colosseum.com/api/v1/archives/DOCUMENT_ID?maxChars=8000"offsetandnextOffsetfor long documents. - Synthesize — After gathering data, provide:
- Top 5 most similar existing projects (with why they're similar)
- Gap analysis summary: what winners emphasize, what they skip
- Crowdedness assessment for the user's niche
- Recommended differentiation angle based on gaps
- Relevant research sources
Non-Negotiables
- Always verify the token before making API calls.
- Use a 15-second timeout on all API requests. If the API is down, tell the user plainly and suggest trying later.
- Present data honestly. Do not spin crowded spaces as "opportunity-rich" or empty spaces as "blue ocean" without evidence.
- If search returns 0 results, try broadening the query before concluding there's nothing relevant.
- Always attribute data to Colosseum Copilot — users should know where the intelligence comes from.
- Do not store or log the user's token anywhere except
~/.superstack/config.json.
Integration with Other Skills
This skill provides data that enriches other idea-phase skills:
/find-next-crypto-idea— Copilot data can validate whether an idea has been tried before and what gaps exist/competitive-landscape— Copilot's project search directly complements ecosystem catalog searches/validate-idea— Winner patterns from Copilot strengthen demand signal analysis
Users can invoke this skill standalone or as a data source during those workflows (only when they explicitly ask or opt in).
Phase Handoff
This skill is Phase 1 (Idea) in the Idea → Build → Launch journey. After analysis:
- If working within a project, write/update
.superstack/idea-context.mdwith acopilot_landscapefield containing:similar_projects: array of top matches with name, one-liner, similarity, hackathongap_analysis: overindexed and underindexed attributes from winner comparisoncrowdedness: assessment of the nicheresearch_sources: relevant archive results
- See
../../../data/specs/phase-handoff.mdfor the full JSON contract.
Quick Start
# Search for similar hackathon projects:
# "Search Colosseum for agent payment projects"
# "What hackathon projects exist for DeFi lending on Solana?"
# Explore categories:
# "Show me the hottest Colosseum clusters"
# "Which hackathon categories have the highest win rate?"
# Gap analysis:
# "What do Solana hackathon winners do differently?"
# "What patterns separate winners from losers?"
# Research:
# "Find research about MEV on Solana"Telemetry (run last)
After the skill workflow completes (success, error, or abort), log the telemetry event.
Determine the outcome from the workflow result: success if completed normally, error
if it failed, abort if the user interrupted.
Run this bash:
_TEL_END=$(date +%s)
_TEL_DUR=$(( _TEL_END - ${_TEL_START:-$_TEL_END} ))
_TEL_TIER=$(cat ~/.superstack/config.json 2>/dev/null | grep -o '"telemetryTier": *"[^"]*"' | head -1 | sed 's/.*"telemetryTier": *"//;s/"$//' || echo "anonymous")
if [ "$_TEL_TIER" != "off" ]; then
echo '{"skill":"colosseum-copilot","phase":"idea","event":"completed","outcome":"OUTCOME","duration_s":"'"$_TEL_DUR"'","session":"'"$_SESSION_ID"'","ts":"'$(date -u +%Y-%m-%dT%H:%M:%SZ)'","platform":"'$(uname -s)-$(uname -m)'"}' >> ~/.superstack/telemetry.jsonl 2>/dev/null || true
true
fiReplace OUTCOME with success/error/abort based on the workflow result.
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.

