Index of every page on this site: llms.txt
All skills
binance avatar

/meme-rush

@7d1b1f7
by binancebinance/binance-skills-hub1.2k stars
281
Run it yourself

Live launchpad feed + AI hot topics for meme tokens. (1) meme-rush: real-time lifecycle feed on launchpads across Solana, BSC and Base (Pump.fun, Four.meme, Flap, ...) — brand-new launches, currently-finalizing / bonding-curve tokens, and just-migrated-to-DEX tokens; filter by dev behavior, age, market cap, and launchpad sub-type (e.g. Open Four Agent ID exclusive launches — anyone can launch; on the bonding curve only Agent ID holders can buy). (2) topic-rush: AI-detected hot market narratives with the associated tokens ranked by inflow. Use for: "new pump.fun launches", "what just migrated", "currently bonding", "hot narratives", "what topic is pumping right now", "live launchpad feed", "AI hot topics", "Agent ID exclusive launches", "Agent ID exclusive bonding curve", "Open Four".

Use this Skill: https://skilld.dev/gh/binance/binance-skills-hub/meme-rush

Nothing lands on disk. Nothing to clean up.

Fork this Skill

Edit a local copy. It keeps the author and licence.

SKILL.md

≈202 tokens for metadata: the name and description. ≈2.3k when used: this file. ≈5.5k more on demand in 1 file.

Description uses 10.1% of example budget

Before choosing a Skill, your Agent reads its name and description. All available Skills share that space.

  • A shorter description leaves more room for other Skills. This entry exceeds our 1% size suggestion.

In our Claude Code example, all Skill names and descriptions share 8,000 characters. This Skill uses ≈806 characters, or 10.1%.

The 1% threshold is a size suggestion. Longer descriptions can still fit.

Your model, settings, and other Skills decide how much text your Agent can read.

Example settings and source

The example uses a 200k-token context and default Claude Code settings. The count includes the name, description, separators, and when_to_use when present. Codex also counts local file paths.

Skit's source and limits: Codex 0.160.1, Claude Code 2.1.292.

Meme Rush Skill

Overview

Two rank feeds fronted by one CLI: meme-rush (launchpad lifecycle tracking) and topic-rush (AI hot-topic discovery with associated tokens). The CLI owns URL, method, JSON encoding, timeout, and upstream error mapping — the agent only picks the subcommand and fills the filter JSON.

When to Use This Skill

User intent Command
New / finalizing / migrated meme tokens on a launchpad meme-rush
AI-generated market hot topics and their associated tokens topic-rush

Supported Chains

Chain chainId Supported on
BSC 56 meme-rush, topic-rush
Solana CT_501 meme-rush, topic-rush
Base 8453 meme-rush

How to Call APIs

node <skill-dir>/scripts/cli.mjs meme-rush '{"chainId":"CT_501","rankType":10,"limit":20}'
node <skill-dir>/scripts/cli.mjs topic-rush '{"chainId":"CT_501","rankType":10,"sort":10,"asc":false}'

Preset: Agent ID exclusive launches

Ready-to-run queries for Open Four Agent ID exclusive launches (openfour.meme.AgentIdExclusive). Use them as-is, and pick the stage from how the user phrased the request; read references/cli.md only when you need other filters or field details.

# New — just launched, not migrated yet (buying requires an Agent ID)
node <skill-dir>/scripts/cli.mjs meme-rush '{"chainId":"56","rankType":10,"protocol":[2004],"tokenVersionFilter":{"2004":["openfour.meme.AgentIdExclusive"]},"limit":50}'
# Finalizing — bonding curve nearly complete, not migrated yet (buying requires an Agent ID)
node <skill-dir>/scripts/cli.mjs meme-rush '{"chainId":"56","rankType":20,"protocol":[2004],"tokenVersionFilter":{"2004":["openfour.meme.AgentIdExclusive"]},"limit":50}'
# Migrated — on a DEX, any address can trade
node <skill-dir>/scripts/cli.mjs meme-rush '{"chainId":"56","rankType":30,"protocol":[2004],"tokenVersionFilter":{"2004":["openfour.meme.AgentIdExclusive"]},"limit":50}'
Parameter Value Notes
chainId "56" Open Four is BSC-only
protocol + tokenVersionFilter as above Both required. Not exclusive: 1 (that is the unrelated "Binance exclusive" flag)
rankType 10 / 20 / 30 No default — choose from the user's wording
limit 50 Max 100
Anything else omit Omitting pairAnchorFilter = all quote tokens. Avoid marketCap / holders filters — brand-new tokens have them null
  • An empty list means no such token exists right now — not an error. Do not silently swap in another sub-type.

Commands

Command Purpose Required args Example
meme-rush Launchpad token lifecycle ranking (new / finalizing / migrated) chainId, rankType node <skill-dir>/scripts/cli.mjs meme-rush '{"chainId":"CT_501","rankType":10,"limit":20}'
topic-rush AI-generated hot topics with associated tokens chainId, rankType, sort node <skill-dir>/scripts/cli.mjs topic-rush '{"chainId":"CT_501","rankType":10,"sort":10}'

Optional filters for meme-rush (all min/max pairs): progress, tokenAge (minutes), holders, liquidity, volume, marketCap, netBuy, count{,Buy,Sell}, holders{Top10,Dev,Sniper,Insider}Percent, bundlerHoldingPercent, newWalletHoldingPercent, bnHoldingPercent, {bn,kol,pro,smartMoney}Holders, devMigrateCount, taxRateBuy, taxRateSell, globalFee; plus keywords, excludes, limit (max 100), protocol[], devPosition, devBurnedToken, excludeDevWashTrading, excludeInsiderWashTrading, exclusive, paidOnDexScreener, pumpfunLiving, cmcBoost, pairAnchorFilter, globalFeeFilter, tokenVersionFilter, creatorType[], excludeCreatorType[], tokenSocials.atLeastOne, tokenSocials.socials[]. See references/cli.md for type and semantics of each field.

Optional filters for topic-rush: asc (boolean), keywords, topicType, tokenSizeMin/Max, netInflowMin/Max.

Rules

  • meme-rush rankType enum — stage of the token's launchpad lifecycle:
    • 10 = New (freshly created, still on bonding curve)
    • 20 = Finalizing (bonding curve nearly complete, about to migrate)
    • 30 = Migrated (just migrated to DEX)
  • topic-rush rankType enum — topic freshness:
    • 10 = Latest (newest hot topics)
    • 20 = Rising (rising topics, all-time-high inflow between $1k–$20k)
  • topic-rush sort enum: 10 = create time, 20 = net inflow. Default to sort=10 when the user does not specify a sort preference.
  • Only chainId and rankType are required for meme-rush; all other parameters are optional filters. topic-rush additionally requires sort.
  • Unknown or misspelled filters are silently ignored — the API returns an unfiltered list with no error. Price (priceMin/Max, priceChangeMin/Max), timestamps (createTime*), sniperCountMin/Max, devSellPercentMin/Max and top-level pairAnchorAddress do NOT filter. For time use tokenAge*; for quote tokens use pairAnchorFilter. Price, price change, sniper count and dev sell % have no equivalent server-side filter: filter the returned rows locally (max 100 per call, so results are not exhaustive), or — only after telling the user the condition changes — use a nearby filter such as marketCap*. See "Filters that look valid but are silently ignored" in references/cli.md.
  • tokenAge is in minutes, not seconds — "launched in the last 5 minutes" is tokenAgeMax: 5; on Migrated (rankType=30) it counts from migrateTime. It replaces a time condition only when the time base matches — for creation time on Migrated, or any other timestamp, filter the returned rows locally and say the result is a sample (max 100).
  • A missing value is not filtered consistently — null tax passes a tax ceiling, null market fields fail both bounds, and null devPosition passes devPosition: 2. A filter meant to avoid something may not enforce it; see "How null behaves" in references/cli.md.
  • null risk tags (tag*) mean "not evaluated", not "safe" — to exclude risky tokens use excludeDevWashTrading: 1 / excludeInsiderWashTrading: 1, never "tag != 1". These excludes only drop tokens already tagged; untagged tokens remain, so the result is still not a safety whitelist. On rankType=10, do not add marketCap / holders / liquidity filters: brand-new tokens have them null and get filtered out.
  • symbol is not unique — identify tokens by (chainId, contractAddress); the address alone is only unique within a single chain.
  • Quote tokens: omit pairAnchorFilter to get all of them (configured quote tokens + Others + RWA). Passing it replaces that default — only what you list is returned, e.g. {"56":{"pairAnchorAddress":["<USDT>"]}} returns USDT-paired tokens only. Each queried chain needs its own key with a non-empty pairAnchorAddress or otherPairAnchor: 1 / rwaPairAnchor: 1; an empty entry or a missing chain key returns an empty list. See pairAnchorFilter in references/cli.md.
  • Percentage fields are pre-formatted — progress, holder %, devSellPercent, taxRateBuy / taxRateSell, priceChange, priceChange24h are already strings like "42.5", so append % directly when displaying; do NOT multiply by 100.
  • Icon URL prefix: icon is a relative path returned by upstream; prepend https://bin.bnbstatic.com before rendering. tokenList[].icon in topic-rush responses follows the same rule.
  • Tax visibility: there is no taxRate field — use taxRateBuy / taxRateSell. A null tax is treated as 0 by the server filter, so unknown-tax tokens pass taxRateBuyMax / taxRateSellMax; show null as "unknown", never 0% or "low tax". For protocol=2001 (Four.meme) they only appear on the Migrated list; for protocol=2002 (Flap) they appear on all lists.
  • Protocol codes: 1xxx = Solana, 2xxx = BSC, 3xxx = Base (e.g. 1001 Pump.fun, 2001 Four.meme, 2002 Flap, 2004 Open Four). Many protocols have little Migrated data — query New / Finalizing too. See references/cli.md for the full table.
  • Open Four sub-types all use protocol=2004 — select them with tokenVersionFilter (for Agent ID exclusive launches use the "Preset: Agent ID exclusive launches" section above). Always pass protocol:[2004] together, otherwise other protocols' tokens are returned too.
    • "Agent ID exclusive" / "Agent ID exclusive launches" means this tokenVersionFilter — not exclusive: 1, which is the unrelated "Binance exclusive" flag.
    • An empty list means no token of that sub-type exists right now — not an error. Say so, and optionally show other Open Four sub-types (query protocol:[2004] without the filter).

Full CLI Reference

See references/cli.md for per-subcommand invocations, full parameter tables (all filter fields, holder-distribution filters, dev & launch filters), return-field tables (core, trade counts, holder distribution, dev & migration, tags & flags, social links, AI narrative, topic + tokenList), and real response samples.

Source: SKILL.md on GitHub

skilld matched fixed text patterns in SKILL.md and file names. Patterns miss obfuscated code.

skilld run checks every file with the same patterns. It asks for approval before it loads a Skill with a behavior marked Needs approval.

1 warning7d5 checks · Risk SAFE
  • Gen Agent Trust Hub8d

    This skill is safe for use and integrates with Binance's official Web3 and market data APIs. It evaluates low-risk for indirect prompt injection because it processes dynamic market data and AI-generated summaries that could theoretically contain instructions to influence the agent.

  • Socket7d

    No alerts

  • Snyk8d

    Risk: MEDIUM · 1 issue

  • Runlayer8mo

    1/1 file flagged

  • ZeroLeaks7mo

    Score: 93/100 · 2 sections analyzed

Signed by skilld at 7d1b1f7. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub last week.

Activeupdated last week
metadata
{
  "author": "binance-web3-team",
  "version": "2.1"
}

README badge

README badge for binance/binance-skills-hub/meme-rush