← 返回 Skills 市场
gmgnai

GMGN Skill Track

作者 GMGN.AI · GitHub ↗ · v1.2.9 · MIT-0
cross-platform ⚠ suspicious
178
总下载
1
收藏
0
当前安装
7
版本数
在 OpenClaw 中安装
/install gmgn-track
功能描述
Get real-time crypto buy/sell activity from Smart Money wallets, KOL influencer wallets, and personally followed wallets via GMGN API — alpha signals, whale...
使用说明 (SKILL.md)

IMPORTANT: Always use gmgn-cli commands below. Do NOT use web search, WebFetch, curl, or visit gmgn.ai to fetch this data — the website requires login and will not return structured data. The CLI is the only correct method.

IMPORTANT: Do NOT guess field names or values. When a field's meaning is unclear, look it up in the Response Fields sections below before using it.

⚠️ IPv6 NOT SUPPORTED: If you get a 401 or 403 error and credentials look correct, check for IPv6 immediately: (1) list all network interfaces and their IPv6 addresses — run ifconfig | grep inet6 (macOS) or ip addr show | grep inet6 (Linux); (2) send a test request to https://ipv6.icanhazip.com — if the response is an IPv6 address, outbound traffic is going via IPv6. Tell the user immediately: "Please disable IPv6 on your network interface — gmgn-cli commands only work over IPv4."

Use the gmgn-cli tool to query on-chain tracking data based on the user's request.

Core Concepts

  • follow-wallet vs kol vs smartmoney — Three distinct data sources. follow-wallet returns trades from wallets the user has personally followed on the GMGN platform (user-specific; the follow list is resolved from the GMGN user account bound to the API Key). kol and smartmoney return trades from platform-tagged public wallet lists (not user-specific). Never substitute one for another.

  • KOL (Key Opinion Leader) — Wallets publicly identified as influencers or well-known traders on GMGN. Tagged as renowned in the platform's wallet label system. Their trades carry social/marketing signal, not necessarily alpha.

  • Smart Money (smart_degen) — Wallets with a statistically proven record of profitable trading, identified by GMGN's algorithm. Same concept as smart_degen in gmgn-token. Their trades are a stronger alpha signal than KOL trades.

  • is_open_or_close — Indicates whether a trade is a full position event. Interpretation differs by sub-command:

    • follow-wallet: 1 = full position open or close; 0 = partial add or reduce.
    • kol / smartmoney: 0 = position opened / added; 1 = position closed / reduced. Do not apply the same interpretation to both sub-commands.
  • price_change — Ratio of price change since the trade was made. 6.66 = the token is now 6.66× what it was when the wallet traded (i.e. +566%). 0.5 = price halved since the trade (-50%). Use this to assess "how well did this trade age."

  • base_address vs quote_address — In a trading pair, base_address is the token being bought/sold; quote_address is what it was priced in (typically SOL native address on Solana). To get the token of interest, always read base_address.

  • maker_info.tags — Array of platform labels on the wallet (e.g. ["kol", "gmgn"], ["smart_degen", "photon"]). A wallet can carry multiple tags. Use tag_rank (follow-wallet only) to see the wallet's rank within each tag category.

  • Cluster signal — When multiple followed/tracked wallets trade the same token in the same direction within a short time window, this is a stronger conviction signal than a single wallet. Highlight this pattern when it appears in results.

When to use which sub-command:

  • track follow-wallet — user asks "what did the wallets I follow trade?", "show me my follow list trades", "show my followed wallet activity" → requires wallets followed via GMGN platform
  • track kol — user asks "what are KOLs buying?", "show me influencer trades", "what are KOLs doing recently" → returns trades from known KOL wallets
  • track smartmoney — user asks "what is smart money doing?", "show me whale trades", "what is smart money buying recently" → returns trades from smart money / whale wallets

Do NOT confuse these three:

  • follow-wallet = wallets the user has personally followed on GMGN
  • kol = platform-tagged KOL / influencer wallets (not user-specific)
  • smartmoney = platform-tagged smart money / whale wallets (not user-specific)

Sub-commands

Sub-command Description
track follow-wallet Trade records from wallets the user personally follows on GMGN
track kol Real-time trades from KOL / influencer wallets tagged by GMGN
track smartmoney Real-time trades from smart money / whale wallets tagged by GMGN

Supported Chains

sol / bsc / base / eth

Prerequisites

  • gmgn-cli installed globally — if missing, run: npm install -g gmgn-cli
  • GMGN_API_KEY configured in ~/.config/gmgn/.env — required for all sub-commands
  • GMGN_PRIVATE_KEY — required for track follow-wallet only (critical auth); not needed for kol or smartmoney

Rate Limit Handling

All tracking routes used by this skill go through GMGN's leaky-bucket limiter with rate=10 and capacity=10. Sustained throughput is roughly 10 ÷ weight requests/second, and the max burst is roughly floor(10 ÷ weight) when the bucket is full.

Command Route Weight
track follow-wallet GET /v1/trade/follow_wallet 3
track kol GET /v1/user/kol 1
track smartmoney GET /v1/user/smartmoney 1

When a request returns 429:

  • Read X-RateLimit-Reset from the response headers. It is a Unix timestamp in seconds that marks when the limit is expected to reset.
  • If the response body contains reset_at (e.g., {"code":429,"error":"RATE_LIMIT_BANNED","message":"...","reset_at":1775184222}), extract reset_at — it is the Unix timestamp when the ban lifts (typically 5 minutes). Convert to local time and tell the user exactly when they can retry.
  • The CLI may wait and retry once automatically when the remaining cooldown is short. If it still fails, stop and tell the user the exact retry time instead of sending more requests.
  • For RATE_LIMIT_EXCEEDED or RATE_LIMIT_BANNED, repeated requests during the cooldown can extend the ban by 5 seconds each time, up to 5 minutes. Do not spam retries.

First-time setup (if GMGN_API_KEY is not configured):

  1. Generate key pair and show the public key to the user:

    openssl genpkey -algorithm ed25519 -out /tmp/gmgn_private.pem 2>/dev/null && \
      openssl pkey -in /tmp/gmgn_private.pem -pubout 2>/dev/null
    

    Tell the user: "This is your Ed25519 public key. Go to https://gmgn.ai/ai, paste it into the API key creation form, then send me the API Key value shown on the page."

  2. Wait for the user's API key, then configure (saves both API key and private key — private key is required for track follow-wallet):

    mkdir -p ~/.config/gmgn
    echo "GMGN_API_KEY=\x3Ckey_from_user>" > ~/.config/gmgn/.env
    echo "GMGN_PRIVATE_KEY=$(awk '{printf "%s\\
    

", $0}' /tmp/gmgn_private.pem)" >> ~/.config/gmgn/.env chmod 600 ~/.config/gmgn/.env rm /tmp/gmgn_private.pem


## Usage Examples

```bash
# Follow-wallet trades (all wallets you follow)
gmgn-cli track follow-wallet --chain sol

# Follow-wallet trades filtered by wallet
gmgn-cli track follow-wallet --chain sol --wallet \x3Cwallet_address>

# Follow-wallet filtered by trade direction
gmgn-cli track follow-wallet --chain sol --side buy

# Follow-wallet filtered by USD amount range
gmgn-cli track follow-wallet --chain sol --min-amount-usd 100 --max-amount-usd 10000

# KOL trade records (SOL, default)
gmgn-cli track kol --limit 10 --raw

# KOL trade records on SOL, buy only
gmgn-cli track kol --chain sol --side buy --limit 10 --raw

# Smart Money trade records (SOL, default)
gmgn-cli track smartmoney --limit 10 --raw

# Smart Money trade records, sell only
gmgn-cli track smartmoney --chain sol --side sell --limit 10 --raw

track follow-wallet Options

Option Description
--chain Required. sol / bsc / base / eth
--wallet \x3Caddress> Filter by wallet address
--limit \x3Cn> Page size (1–100, default 10)
--side \x3Cside> Trade direction: buy / sell
--filter \x3Ctag...> Repeatable filter conditions
--min-amount-usd \x3Cn> Minimum trade amount (USD)
--max-amount-usd \x3Cn> Maximum trade amount (USD)

track kol / track smartmoney Options

Option Description
--chain \x3Cchain> Required. Chain: sol / bsc / base / eth
--limit \x3Cn> Page size (1–200, default 100)
--side \x3Cside> Filter by trade direction: buy / sell (client-side filter — applied locally after fetching results)

track follow-wallet Response Fields

Top-level fields:

Field Description
next_page_token Opaque token for fetching the next page of results
list Array of trade records

Each item in list contains:

Field Description
id Record ID (base64-encoded, use as cursor)
chain Chain name (e.g. sol)
transaction_hash On-chain transaction hash
maker Wallet address of the followed wallet
side Trade direction: buy or sell
base_address Token contract address
quote_address Quote token address (SOL native address for buys/sells on SOL)
base_amount Token quantity in smallest unit
quote_amount Quote token amount spent / received (e.g. SOL)
amount_usd Trade value in USD
cost_usd Same as amount_usd — USD value of this transaction leg
buy_cost_usd Original buy cost in USD (0 if this record is the buy itself)
price Token price denominated in quote token at time of trade
price_usd Token price in USD at time of trade
price_now Token current price in USD
price_change Price change ratio since trade time (e.g. 6.66 = +666%)
timestamp Unix timestamp of the trade
is_open_or_close 1 = full position open or close; 0 = partial add or reduce
launchpad Launchpad display name (e.g. Pump.fun)
launchpad_platform Launchpad platform identifier (e.g. Pump.fun, pump_agent)
migrated_pool_exchange DEX the token migrated to, if any (e.g. pump_amm); empty if not migrated
base_token.symbol Token ticker symbol
base_token.logo Token logo image URL
base_token.hot_level Hotness level (0 = normal, higher = trending)
base_token.total_supply Total token supply (string)
base_token.token_create_time Unix timestamp when token was created
base_token.token_open_time Unix timestamp when trading opened (0 if not yet migrated/opened)
maker_info.address Followed wallet address
maker_info.name Wallet display name
maker_info.twitter_username Twitter / X username
maker_info.twitter_name Twitter / X display name
maker_info.tags Array of wallet tags (e.g. ["kol","gmgn"])
maker_info.tag_rank Map of tag → rank within that category (e.g. {"kol": 854})
balance_info Wallet token balance info; null if not available

track kol / track smartmoney Response Fields

The response is an object with a list array. Each item in list contains:

Field Description
transaction_hash On-chain transaction hash
maker Wallet address of the trader (KOL / Smart Money)
side Trade direction: buy or sell
base_address Token contract address
base_token.symbol Token ticker symbol
base_token.launchpad Launchpad platform (e.g. pump)
amount_usd Trade value in USD
token_amount Token quantity traded
price_usd Token price in USD at time of trade
buy_cost_usd Original buy cost in USD (0 if this record is the buy)
is_open_or_close 0 = position opened / added, 1 = position closed / reduced
timestamp Unix timestamp of the trade
maker_info.twitter_username KOL's Twitter username
maker_info.tags Wallet tags (e.g. kol, smart_degen, photon)

Smart Money Behavior Interpretation

After receiving trade data, interpret the signals using these frameworks before presenting results. Do not just list trades — analyze what they mean.

1. Signal Strength Levels

Level Criteria
Weak 1 KOL buys
Medium 2–3 smart money buys in the same direction, OR 1 smart money full position open
Strong ≥ 3 smart money wallets same direction within 30 min (cluster signal)
Very Strong Cluster signal + full position opens + KOL joining the same trade

2. Reading is_open_or_close — Conviction Signals

The field has opposite meanings by sub-command:

  • follow-wallet: 1 = full position open or close; 0 = partial add or reduce.
  • kol / smartmoney: 0 = position opened / added; 1 = position closed / reduced.

Full position events (full open or full close) carry much stronger conviction than partial adds. A wallet opening a full new position signals high confidence. A wallet doing a full close signals they are exiting completely — treat this as a potential exit signal for that token.

3. Using price_change to Evaluate Track Record

price_change is a ratio of current price vs price at trade time:

  • price_change > 2 → this wallet's trade aged well (token is now 2x+ since they bought) — strong conviction signal
  • price_change 1–2 → modest gain, trade is in profit
  • price_change \x3C 1 → trade is underwater (current price below entry)

Use this to build a mental model of a wallet's past performance before acting on their current trades.

4. Cluster Signal Detection

When multiple trades hit the same base_address in a short time window, this is a convergence signal — stronger than any single trade. To identify:

  • Group results by base_address
  • Count distinct maker addresses trading the same direction
  • If ≥ 3 distinct wallets buy the same token within ~30 min → highlight as cluster signal

Cluster signals from smartmoney are stronger than from kol alone.

5. Red Flags in Smart Money Data

  • Smart money selling (side = sell + is_open_or_close = full close) → exit signal — evaluate whether to exit or reduce position
  • Only KOL buying, zero smart_degen → social hype without fundamental backing; higher risk
  • Renowned buying + smart money selling simultaneously → divergence signal — insiders may be distributing into retail/KOL demand; high risk
  • Single very large buy, no follow-through → may be one-off; wait for confirmation from other wallets

Output Format

track follow-wallet / track kol / track smartmoney — Trade Feed

Present as a reverse-chronological trade feed. Do not dump raw JSON.

{timestamp}  {side}  {base_token.symbol}  ${amount_usd}  by {maker_info.name or short address}
             [{tags}]  Price: ${price_usd}  |  Price now: ${price_now}  ({price_change}x since trade)

Group by token if multiple trades hit the same token. Highlight tokens where several followed wallets traded in the same direction within a short window (cluster signal).

For follow-wallet, also show is_open_or_close: flag full position opens/closes distinctly from partial adds/reduces.

Cluster Signal Summary

After presenting the trade feed, check for convergence signals. If ≥ 2 distinct wallets traded the same token in the same direction, display a summary block:

⚡ Convergence Signals
──────────────────────────────────────────
TOKEN_X ({short_address})
  5 smart money wallets — all BUY — $42,300 total — within 15 min
  Signal strength: STRONG

TOKEN_Y ({short_address})
  2 KOL wallets — BUY (full open) — $8,100 total
  Signal strength: MEDIUM

For STRONG signals: proceed to full token research before acting — see docs/workflow-token-research.md For MEDIUM signals: monitor and wait for more wallets to confirm before acting.

If no convergence signals are detected: output "No cluster signals detected in this result set."

To research any token surfaced by smart money activity, follow docs/workflow-token-research.md

Smart money leaderboard / wallet profiling: When the user asks "which smart money wallets are best to follow", "rank wallets by win rate", or wants to compare wallet performance — use track smartmoney to collect active wallet addresses, then batch-query their stats via gmgn-portfolio stats. Full workflow: docs/workflow-smart-money-profile.md

Daily brief: When the user asks for a market overview ("what's the market like today", "what is smart money buying today", "give me a daily brief") — combine track smartmoney + track kol with gmgn-market trending. Full workflow: docs/workflow-daily-brief.md

Safety Constraints

  • follow-wallet reveals your following list — results expose which wallets you have followed on GMGN. Do not share raw output in public channels.
  • track kol / track smartmoney expose no personal data — these use API Key auth only and return platform-tagged public wallet activity. Safe to share raw output.

Notes

  • track follow-wallet uses critical auth (API Key + private key signature); track kol and track smartmoney use normal auth (API Key only)
  • track follow-wallet returns trades from wallets followed on the GMGN platform; the follow list is resolved automatically from the GMGN user account bound to the API Key — --wallet is optional
  • Use --raw to get single-line JSON for further processing
  • track kol / track smartmoney --side is a client-side filter — the CLI fetches all results then filters locally; it is NOT sent to the API
安全使用建议
Before installing or using this skill: (1) do not trust the registry metadata here — SKILL.md requires GMGN_API_KEY and GMGN_PRIVATE_KEY and a gmgn-cli npm package even though the metadata lists none; ask the publisher to correct the metadata. (2) Treat GMGN_PRIVATE_KEY as highly sensitive — prefer an API token with limited scope if possible and avoid placing private keys in plaintext home-config files. (3) Verify the origin and source code of the gmgn-cli npm package (npm page, repository, maintainer, recent publish history, downloads, and whether the package executes code at install). Consider running it in a sandbox/container or reviewing its code before global installation. (4) Be aware the skill asks you to run system network commands (ifconfig/ip) and an external IP-check; those are troubleshooting steps but will reveal local network info. (5) If you must use follow-wallet features, minimize blast radius by creating limited-scope credentials and avoid sharing other unrelated credentials. (6) Ask the skill author to add explicit install metadata and declare required env vars/config paths so you can make an informed decision.
功能分析
Type: OpenClaw Skill Name: gmgn-track Version: 1.2.9 The skill bundle contains instructions in SKILL.md that direct the AI agent to perform system reconnaissance by executing 'ifconfig' or 'ip addr' to check for IPv6 addresses, and to use 'curl' against 'ipv6.icanhazip.com'. It also involves a setup process where the agent generates an Ed25519 private key using 'openssl' and stores it in '~/.config/gmgn/.env'. While these actions are contextually justified for troubleshooting network connectivity and configuring signed API requests for the 'gmgn-cli' tool, the combination of system-level information gathering and sensitive credential management represents a high-risk profile.
能力标签
cryptorequires-walletcan-sign-transactionsrequires-sensitive-credentials
能力评估
Purpose & Capability
The SKILL.md clearly requires gmgn-cli and user credentials (GMGN_API_KEY and GMGN_PRIVATE_KEY for follow-wallet) and a config path (~/.config/gmgn/.env). The registry metadata lists no required env vars, no primary credential, and no install spec. That mismatch (claimed 'none' vs documented required credentials/install) is incoherent: the skill legitimately needs those credentials for its functionality, but the metadata does not declare them.
Instruction Scope
Runtime instructions direct the agent to run gmgn-cli for all queries (appropriate) but also to read the user's GMGN config file and to run system network commands (ifconfig/ip addr) and an external IP test (https://ipv6.icanhazip.com) to debug IPv6. Reading the user's GMGN config and private key is necessary for follow-wallet functionality, but those file/env accesses are not declared in metadata. The guidance to never use webfetch/curl for GMGN data but to use an external endpoint for IPv6 testing is odd but explainable as a connectivity check.
Install Mechanism
There is no install spec in the registry entry, but SKILL.md instructs users to run `npm install -g gmgn-cli` if gmgn-cli is missing. Installing a third-party npm package globally is a moderate-risk action (npm packages can execute code during install); this is plausible for a CLI-based skill but would be safer and clearer if the registry install spec declared the package and its source (npm package page / homepage / repo).
Credentials
The skill requires GMGN_API_KEY and (for follow-wallet) a GMGN_PRIVATE_KEY stored in ~/.config/gmgn/.env according to SKILL.md — both are sensitive. These credentials are proportionate to the stated, user-specific follow-wallet capability, but the registry metadata lists no required env vars or primary credential, which is a notable omission. Requiring a private key increases sensitivity; the skill should clearly document why both API key and private key are needed and how they are stored/used.
Persistence & Privilege
The skill is instruction-only, has no install spec, does not request 'always: true', and does not declare the ability to modify other skills or system-wide settings. Autonomous invocation is allowed (default) but not combined with other high-risk factors here.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install gmgn-track
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /gmgn-track 触发
  4. 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.2.9
gmgn-track 1.2.9 - Added support for tracking on the Ethereum (eth) chain in addition to Solana, BSC, and Base. - Updated usage instructions and argument hints to include "eth" as a supported chain. - Clarified that GMGN_PRIVATE_KEY is now required only for `track follow-wallet` (not for kol/smartmoney). - Improved first-time setup instructions to save both API key and private key for follow-wallet tracking. - Minor improvements to documentation and descriptions for clarity.
v1.2.8
gmgn-track 1.2.8 changelog - Updated the skill description for clearer, more concise messaging focused on copy-trading, alpha, and whale tracking use cases. - Clarified and shortened the audience definitions and scope in the introductory section. - Made usage scenarios and command guidance more straightforward. - No internal logic or functional changes were made.
v1.2.7
gmgn-track 1.2.7 changelog - Updated description for greater clarity, specifying primary use cases and referring users to gmgn-portfolio for deep wallet analysis. - Improved detail on sub-command intent and data separation between follow-wallet, KOL, and smartmoney sources. - Enhanced guidance on when to use each sub-command and on how to interpret key fields. - General text refinements for easier understanding of concepts and usage. - No functional or file changes; documentation-only update.
v1.2.6
gmgn-track 1.2.5 - Updated skill description for improved clarity: "crypto token trading activity" replaces "trade activity". - No changes to functionality or CLI usage. - No file changes detected; documentation only update.
v1.2.5
gmgn-track 1.2.5 - Updated documentation for clarity: clarified that an API Key is required for all sub-commands and a private key is no longer needed for "follow-wallet" calls. - Corrected description of "follow-wallet" logic: follow list is now user-specific and resolved from the API Key, not a private key. - Improved and condensed skill description to highlight use cases and supported blockchains. - Removed outdated first-time setup instructions relating to private key; first-time API key setup now requires only the API Key. - No functional or interface changes—documentation and guidance only.
v1.2.3
gmgn-track 1.2.3 Changelog - Added explicit documentation for rate limit handling, including cooldown instructions and `reset_at` handling for 429 errors. - Clarified that all tracking routes use a leaky-bucket limiter, listing exact command weights and burst/throughput constraints. - Reduced and streamlined the description, emphasizing major sub-commands and removing extensive field tables and verbose metadata. - Updated metadata to include `cliHelp` field. - No changes to core CLI usage or command structure.
v1.1.9
Initial release. Track smart money, KOL wallets, and followed wallets on Solana, BNB Chain, and Base Chain — real-time on-chain trade signals and copy trading insights for meme coins, pump.fun and four.meme tokens via GMGN.
元数据
Slug gmgn-track
版本 1.2.9
许可证 MIT-0
累计安装 0
当前安装数 0
历史版本数 7
常见问题

GMGN Skill Track 是什么?

Get real-time crypto buy/sell activity from Smart Money wallets, KOL influencer wallets, and personally followed wallets via GMGN API — alpha signals, whale... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 178 次。

如何安装 GMGN Skill Track?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install gmgn-track」即可一键安装,无需额外配置。

GMGN Skill Track 是免费的吗?

是的,GMGN Skill Track 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。

GMGN Skill Track 支持哪些平台?

GMGN Skill Track 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。

谁开发了 GMGN Skill Track?

由 GMGN.AI(@gmgnai)开发并维护,当前版本 v1.2.9。

💬 留言讨论