Monitor Runners

${var} — Filter by chain (e.g. "solana", "eth", "base", "bsc"). If empty, scans all major networks.

Read memory/MEMORY.md for context. Read the last 2 days of memory/logs/ to see if any of these tokens were flagged before (repeat runners are interesting).

Data Source

GeckoTerminal API (free, no API key). Docs: https://api.geckoterminal.com/api/v2

Key endpoints:

• GET /networks/trending_pools?page=1 — trending pools across all networks
• GET /networks/{network}/trending_pools?page=1 — per-network trending pools
• GET /networks/{network}/new_pools?page=1 — newly created pools

Each pool object includes:

• attributes.name — pool name (e.g. "TOKEN / SOL")
• attributes.price_change_percentage.{m5,m15,m30,h1,h6,h24} — price changes
• attributes.volume_usd.{m5,m15,m30,h1,h6,h24} — volume
• attributes.market_cap_usd / attributes.fdv_usd — market cap
• attributes.transactions.h24.{buys,sells} — transaction counts
• attributes.pool_created_at — pool creation timestamp
• attributes.reserve_in_usd — liquidity
• relationships.network.data.id — chain name

Steps

1. Fetch trending pools from multiple networks

Pull trending pools from global + individual chains sequentially:

TMPDIR=$(mktemp -d)

# Fetch sequentially with 1s delay to avoid GeckoTerminal 429 rate limits
for NETWORK in "" solana eth base bsc arbitrum; do
  if [ -z "$NETWORK" ]; then
    URL="https://api.geckoterminal.com/api/v2/networks/trending_pools?page=1"
    OUT="$TMPDIR/global.json"
  else
    URL="https://api.geckoterminal.com/api/v2/networks/${NETWORK}/trending_pools?page=1"
    OUT="$TMPDIR/${NETWORK}.json"
  fi
  curl -s "$URL" > "$OUT"
  # If 429, wait 2s and retry once
  if grep -q '"status":"429"' "$OUT" 2>/dev/null; then
    sleep 2
    curl -s "$URL" > "$OUT"
  fi
  sleep 1
done

If ${var} is set to a specific chain, only fetch that chain's trending pools.

Also fetch new pools to catch tokens that just launched and are already running:

curl -s "https://api.geckoterminal.com/api/v2/networks/new_pools?page=1" > "$TMPDIR/new.json"

2. Merge, dedupe, and rank

From all fetched data:

1. Dedupe by pool address — same pool can appear in global + per-network results
2. Filter to positive 24h movers only (we want runners, not dumps)
3. Sort by price_change_percentage.h24 descending
4. Apply quality filters — skip pools that look like obvious rugs or noise:
   • Skip if 24h volume < $50,000 (too thin)
   • Skip if pool created < 1 hour ago and no meaningful volume (too new to judge)
   • Skip if buy/sell ratio is extreme (>10:1 sells = probably dumping)
   • Flag but don't skip if mcap is null/zero (might be very new)

3. Select the top 5 runners

Pick the 5 tokens with the largest 24h price increase that pass the quality filters.

For each runner, collect:

• Token name and pair (e.g. "TOKEN / SOL")
• Chain (solana, eth, base, etc.)
• 24h change — the big number
• 1h change — is it still running or cooling off?
• 5m change — real-time momentum
• 24h volume — how much money flowed through
• Market cap / FDV — size context
• Buy/sell ratio — demand signal
• Pool age — when was the pool created
• Liquidity (reserve_in_usd) — how much is backing it

4. Analyze each runner

For each of the 5, write a quick assessment:

• Momentum — is it accelerating (1h > 0 and 5m > 0) or fading?
• Volume — proportional to mcap? Over-traded is suspicious, under-traded means thin.
• Buy pressure — more buys than sells = continued demand
• Age — brand new (<24h) = speculative; older = something changed
• Risk — low liquidity, no mcap data, extreme moves (>1000%) = high risk

5. Notify

Send via ./notify (under 4000 chars). No leading spaces on any line:

*runners — ${today}*

1. TOKEN (chain) +X,XXX% 24h
vol $Xm | fdv $Xm | still running / cooling off — [1-2 sentence analysis]

2. TOKEN (chain) +XXX% 24h
vol $Xm | fdv $Xm | momentum note — [1-2 sentence analysis]

3. ...
4. ...
5. ...

vibe: [one line on overall market mood]

6. Log

Append to memory/logs/${today}.md:

## Monitor Runners
- **Networks scanned:** N
- **Positive movers found:** N
- **Top 5:**
  1. TOKEN (chain) +X% — $Xm vol, $Xm mcap — [momentum note]
  2. ...
- **Repeat runners from previous days:** [list any or "none"]
- **Notification sent:** yes

If any token appears as a runner on multiple days in a row, flag it in memory/MEMORY.md — sustained momentum is notable.