agentleFS
sendaifunsendaifun/solana-new/skills/idea/colosseum-copilot/SKILL.md

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

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
  1. Very specific:

    If TEL_PROMPTED is no: Before starting the skill workflow, ask the user about telemetry.

    line 26
  2. Distinctive:

    If A: run this bash:

    line 36
  3. Distinctive:

    If B: run this bash:

    line 43
  4. Distinctive:

    Search and analyze Colosseum's database of 5,400+ Solana hackathon projects.

    line 58
  5. Distinctive:

    Before using this skill, check if the user has a Colosseum Copilot token configured.

    line 62
  6. Distinctive:

    Ask the user with AskUserQuestion:

    line 75
  7. Distinctive:

    If A: Ask for the token, then save it:

    line 88
  8. Very specific:

    If {"authenticated": true}, proceed.

    line 110
  9. Distinctive:

    If B: Tell the user they can set it later via:

    line 112
  10. Distinctive:

    Verify the token is still valid:

    line 121
  11. Distinctive:

    If not authenticated, tell the user their token has expired and walk them through the setup above.

    line 126
  12. Distinctive:

    Read references/copilot-api-guide.md for the full API reference.

    line 130
  13. Distinctive:

    Always verify the token before making API calls.

    line 202
  14. Distinctive:

    Use a 15-second timeout on all API requests.

    line 203
  15. Distinctive:

    If search returns 0 results, try broadening the query before concluding there's nothing relevant.

    line 205
  16. Distinctive:

    Always attribute data to Colosseum Copilot — users should know where the intelligence comes from.

    line 206
  17. Very specific:

    Do not store or log the user's token anywhere except ~/.superstack/config.json.

    line 207
  18. Very specific:

    If working within a project, write/update .superstack/idea-context.md with a copilot_landscape field containing:

    line 223
  19. Very specific:

    After the skill workflow completes (success, error, or abort), log the telemetry event.

    line 251

≈ “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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

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
fi

If 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-prompted

If B: run this bash:

echo '{"telemetryTier":"off"}' > ~/.superstack/config.json
_TEL_TIER="off"
touch ~/.superstack/.telemetry-prompted

This 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"
fi

If 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.

  1. Sign up at https://arena.colosseum.org/copilot
  2. 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/status

If {"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_PAT environment variable
  • Or run /colosseum-copilot again

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/status

If 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.

  1. 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.
  2. Search similar projects — When the user has an idea or query:
    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}'
    Present results as a table: name, one-liner, hackathon, similarity score, prize info, crowdedness.
  3. Gap analysis — Compare what winners do vs the field:
    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}'
    Synthesize overindexed (winners focus on) and underindexed (winners skip) attributes. Present the delta clearly.
  4. Explore clusters — Browse the 30 ML-derived project clusters:
    curl -s -H "Authorization: Bearer $_COPILOT_TOKEN" -H "Content-Type: application/json" https://copilot.colosseum.com/api/v1/filters
    Show: hot categories (most projects), winning categories (highest win rate), open categories (least crowded).
  5. Deep dive into a project — When the user wants full details on a specific project:
    curl -s -H "Authorization: Bearer $_COPILOT_TOKEN" -H "Content-Type: application/json" \
      https://copilot.colosseum.com/api/v1/projects/by-slug/PROJECT_SLUG
    Shows: description, team members (GitHub/Twitter), links (demo, presentation, GitHub), full tag set.
  6. Explore a specific cluster — Get cluster summary and representative projects:
    curl -s -H "Authorization: Bearer $_COPILOT_TOKEN" -H "Content-Type: application/json" \
      https://copilot.colosseum.com/api/v1/clusters/CLUSTER_KEY
    Shows: summary, project count, winner count, representative projects, top tags.
  7. Search archives — Find research sources:
    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"}'
    Present: title, author, source, published date, snippet.
  8. Read full archive document — When a snippet is interesting and the user wants more:
    curl -s -H "Authorization: Bearer $_COPILOT_TOKEN" -H "Content-Type: application/json" \
      "https://copilot.colosseum.com/api/v1/archives/DOCUMENT_ID?maxChars=8000"
    Supports pagination via offset and nextOffset for long documents.
  9. 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:

  1. If working within a project, write/update .superstack/idea-context.md with a copilot_landscape field containing:
    • similar_projects: array of top matches with name, one-liner, similarity, hackathon
    • gap_analysis: overindexed and underindexed attributes from winner comparison
    • crowdedness: assessment of the niche
    • research_sources: relevant archive results
  2. See ../../../data/specs/phase-handoff.md for 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
fi

Replace OUTCOME with success/error/abort based on the workflow result.

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.