docs: enrich all SKILL.md files with response fields, usage examples, and AI-agent guidance

- gmgn-token: full response field reference for all 5 sub-commands (token info nested objects pool/link/stat/wallet_tags_stat, security, pool, holders, traders); granular usage examples per scenario; full due diligence workflow with security threshold table; market cap calculation note; volume redirect to kline
- gmgn-market: verified kline/trending response fields from live API; trending + platform filter examples for SOL/BSC/Base major launchpads; interval selection guide; trenches intent→type mapping table; standardized section headings with sub-command names
- gmgn-portfolio/gmgn-swap: standardize section headings to include sub-command names
- CLAUDE.md: add newly-launched token triggers for gmgn-market skill routing
- Readme/Readme.zh: add missing portfolio sub-commands, kline examples, auto-slippage swap example; remove version pin; update install instructions

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
This commit is contained in:
gumponchain
2026-03-26 21:25:38 +08:00
committed by GMGN.AI
parent 12d3c358d8
commit 0239e42b50
7 changed files with 733 additions and 68 deletions
+245 -17
View File
@@ -10,9 +10,9 @@ Use the `gmgn-cli` tool to query K-line data for a token, browse trending tokens
| Sub-command | Description |
|-------------|-------------|
| `market kline` | Token candlestick data |
| `market trending` | Trending token swap data |
| `market trenches` | Trenches token lists (new creation, near completion, completed) |
| `market kline` | Token candlestick / OHLCV data and trading volume over a time range |
| `market trending` | Trending tokens ranked by swap activity — use `--interval` to specify the time window (e.g. `1m` for 1-minute hottest, `1h` for 1-hour trending) |
| `market trenches` | Newly launched launchpad paltform tokens — **use this when the user asks for "new tokens", "just launched tokens", "latest tokens on pump.fun/letsbonk"**. Three categories: `new_creation` (just created), `near_completion` (bonding curve almost full), `completed` (graduated to open market / DEX) |
## Supported Chains
@@ -24,21 +24,50 @@ Use the `gmgn-cli` tool to query K-line data for a token, browse trending tokens
- Run from the directory where your `.env` file is located, or set `GMGN_HOST` in your environment
- `gmgn-cli` installed globally: `npm install -g gmgn-cli@1.1.0`
## Kline Parameters
## `market kline` Parameters
| Parameter | Required | Description |
|-----------|----------|-------------|
| `--chain` | Yes | `sol` / `bsc` / `base` |
| `--address` | Yes | Token contract address |
| `--resolution` | Yes | Candlestick resolution |
| `--resolution` | Yes | Candlestick resolution: `1m` / `5m` / `15m` / `1h` / `4h` / `1d` |
| `--from` | No | Start time (Unix seconds) |
| `--to` | No | End time (Unix seconds) |
## Resolutions
## `market kline` Response Fields
`1m` / `5m` / `15m` / `1h` / `4h` / `1d`
The response is an object with a `list` array. Each element in `list` is one candlestick:
## Trending Options
| Field | Type | Description |
|-------|------|-------------|
| `time` | number | Candle open time — Unix timestamp in **milliseconds** (divide by 1000 for seconds) |
| `open` | string | Opening price in USD at the start of the period |
| `close` | string | Closing price in USD at the end of the period |
| `high` | string | Highest price in USD during the period |
| `low` | string | Lowest price in USD during the period |
| `volume` | string | Trading volume in **USD** (dollar value of all trades in this period) |
| `amount` | string | Trading volume in **base token units** (number of tokens traded) |
**Important distinctions (naming is counterintuitive — do not guess):**
- `volume` = USD dollar value (e.g. `1214` means ~$1,214 traded) — use this for "how much was traded in USD"
- `amount` = token count (e.g. `5379110` means ~5.38M tokens changed hands) — use this for "how many tokens were traded"
- For tokens not priced at $1, `volume` and `amount` will differ by orders of magnitude (e.g. a $0.0002 token: $1,214 volume = 5,379,110 tokens)
- To get **total USD volume over a time range**, sum `volume` across all candles in the range
- To get **price trend**, read `close` values in chronological order (`time` ascending)
- To detect **volatility**, compare `high` vs `low` within each candle
- Candles are returned in chronological order (oldest first)
## `market trending` Options
**`--interval` selection guide — always match to the user's stated time window:**
| User says | `--interval` |
|-----------|-------------|
| "1分钟热门" / "1m trending" / "hottest right now" | `1m` |
| "5分钟" / "5m" | `5m` |
| "1小时" / "1h" / no time specified (default) | `1h` |
| "6小时" / "6h" | `6h` |
| "24小时" / "今日" / "daily" | `24h` |
| Option | Description |
|--------|-------------|
@@ -52,6 +81,8 @@ Use the `gmgn-cli` tool to query K-line data for a token, browse trending tokens
## Usage Examples
### Kline
```bash
# Last 1 hour of 1-minute candles
# macOS:
@@ -73,22 +104,210 @@ gmgn-cli market kline \
--to $(date +%s)
# Linux: use $(date -d '24 hours ago' +%s) instead of $(date -v-24H +%s)
# Raw output for further processing
gmgn-cli market kline --chain sol --address <addr> \
--resolution 5m --from <ts> --to <ts> --raw | jq '.[]'
```
### Trending — General
```bash
# Top 20 hot tokens on SOL in the last 1 hour, sorted by volume
gmgn-cli market trending --chain sol --interval 1h --order-by volume --limit 20
# Top 50 tokens on SOL, 5m window, sorted by volume
gmgn-cli market trending --chain sol --interval 5m --order-by volume --limit 50
# Hot tokens with social links only, verified and not honeypot, on BSC over 24h
gmgn-cli market trending \
--chain bsc --interval 24h \
--filter has_social --filter not_honeypot --filter verified
# Pump.fun platform tokens on SOL, last 6 hours
gmgn-cli market trending --chain sol --interval 6h --platform Pump.fun
# Raw output for further processing
gmgn-cli market kline --chain sol --address <addr> \
--resolution 5m --from <ts> --to <ts> --raw | jq '.[]'
```
### Trending — SOL by Launchpad Platform
Use `--platform` to filter trending results to tokens from specific launchpads only.
```bash
# SOL 1m hottest — Pump.fun + letsbonk only (most active launchpads), sorted by volume
gmgn-cli market trending \
--chain sol --interval 1m \
--platform Pump.fun --platform letsbonk \
--order-by volume --limit 50 --raw
# SOL 5m hottest — Pump.fun + letsbonk + Moonshot, sorted by volume
gmgn-cli market trending \
--chain sol --interval 5m \
--platform Pump.fun --platform letsbonk --platform moonshot_app \
--order-by volume --limit 50 --raw
# SOL 1h trending — Pump.fun only, with safety filters
gmgn-cli market trending \
--chain sol --interval 1h \
--platform Pump.fun \
--filter renounced --filter frozen --filter not_wash_trading \
--order-by volume --limit 20 --raw
# SOL 1h trending — all major launchpads combined
gmgn-cli market trending \
--chain sol --interval 1h \
--platform Pump.fun --platform letsbonk --platform moonshot_app \
--platform pump_mayhem --platform pump_mayhem_agent --platform bonkers \
--order-by volume --limit 50 --raw
```
### Trending — BSC by Launchpad Platform
```bash
# BSC 1m hottest — fourmeme (main BSC launchpad), sorted by volume
gmgn-cli market trending \
--chain bsc --interval 1m \
--platform fourmeme --platform four_xmode_agent \
--order-by volume --limit 50 --raw
# BSC 5m hottest — fourmeme family, sorted by volume
gmgn-cli market trending \
--chain bsc --interval 5m \
--platform fourmeme --platform fourmeme_agent --platform bn_fourmeme --platform four_xmode_agent \
--order-by volume --limit 50 --raw
# BSC 1h trending — fourmeme with safety filters
gmgn-cli market trending \
--chain bsc --interval 1h \
--platform fourmeme --platform fourmeme_agent --platform bn_fourmeme --platform four_xmode_agent \
--filter not_honeypot --filter verified \
--order-by volume --limit 20 --raw
```
### Trending — Base by Launchpad Platform
```bash
# Base 1m hottest — clanker + zora (main Base launchpads), sorted by volume
gmgn-cli market trending \
--chain base --interval 1m \
--platform clanker --platform zora --platform zora_creator \
--order-by volume --limit 50 --raw
# Base 5m hottest — clanker + zora + virtuals_v2 + flaunch, sorted by volume
gmgn-cli market trending \
--chain base --interval 5m \
--platform clanker --platform zora --platform zora_creator \
--platform virtuals_v2 --platform flaunch \
--order-by volume --limit 50 --raw
# Base 1h trending — all major launchpads with safety filters
gmgn-cli market trending \
--chain base --interval 1h \
--platform clanker --platform zora --platform zora_creator \
--platform virtuals_v2 --platform flaunch --platform baseapp \
--filter not_honeypot --filter verified \
--order-by volume --limit 20 --raw
```
## `market trending` Response Fields
The response is `data.rank` — an array of rank items. Each item represents one token.
**Basic Info**
| Field | Description |
|-------|-------------|
| `address` | Token contract address |
| `symbol` / `name` | Token ticker and full name |
| `logo` | Token logo image URL |
| `chain` | Chain identifier |
| `total_supply` | Total token supply |
| `creator` | Creator wallet address |
| `launchpad_platform` | Launch/pool platform (e.g. `Pump.fun`, `letsbonk`, `pool_meteora`, `fourmeme`) |
| `exchange` | Current DEX (e.g. `meteora_damm_v2`, `raydium`, `pump_amm`) |
| `open_timestamp` | Open market listing time (Unix seconds) |
| `creation_timestamp` | Token creation time (Unix seconds) |
| `rank` | Position in this trending list (lower = hotter) |
| `hot_level` | Trending intensity level (higher = hotter) |
**Price & Market**
| Field | Description |
|-------|-------------|
| `price` | Current price in USD |
| `market_cap` | Market cap in USD (directly available — no calculation needed) |
| `liquidity` | Current liquidity in USD |
| `volume` | Trading volume in USD for the queried interval |
| `history_highest_market_cap` | All-time highest market cap in USD |
| `initial_liquidity` | Initial liquidity at token launch |
| `price_change_percent` | Price change % for the queried interval |
| `price_change_percent1m` | Price change % in last 1 minute |
| `price_change_percent5m` | Price change % in last 5 minutes |
| `price_change_percent1h` | Price change % in last 1 hour |
**Trading Activity**
| Field | Description |
|-------|-------------|
| `swaps` | Total swap count in the queried interval |
| `buys` / `sells` | Buy / sell count in the interval |
| `holder_count` | Number of unique token holders |
| `gas_fee` | Average gas fee per transaction |
**Security & Risk**
| Field | Chains | Description |
|-------|--------|-------------|
| `renounced_mint` | SOL | Mint authority renounced (`1` = yes, `0` = no) |
| `renounced_freeze_account` | SOL | Freeze authority renounced (`1` = yes, `0` = no) |
| `is_honeypot` | BSC / Base | Honeypot flag (`1` = yes, `0` = no) |
| `is_open_source` | all | Contract verified (`1` = yes, `0` = no) |
| `is_renounced` | all | Ownership renounced (`1` = yes, `0` = no) |
| `buy_tax` / `sell_tax` | all | Tax rate — empty string means `0` (no tax) |
| `burn_status` | all | Liquidity burn status (e.g. `"none"`, `"burn"`) |
| `top_10_holder_rate` | all | Top 10 wallets concentration (01) |
| `rug_ratio` | all | Rug pull risk score (01) |
| `is_wash_trading` | all | Wash trading detected (`true` / `false`) |
| `rat_trader_amount_rate` | all | Ratio of insider/sneak trading volume |
| `bundler_rate` | all | Ratio of bundle bot trading volume |
| `entrapment_ratio` | all | Entrapment trading ratio |
| `sniper_count` | all | Number of sniper wallets at launch |
| `bot_degen_count` / `bot_degen_rate` | all | Bot degen wallet count / ratio |
| `dev_team_hold_rate` | all | Dev team holding ratio |
| `top70_sniper_hold_rate` | all | Top 70 sniper current holding ratio |
| `lock_percent` | all | Liquidity lock percentage |
**Dev Status**
| Field | Description |
|-------|-------------|
| `creator_token_status` | Dev holding status: `creator_hold` (still holding) / `creator_close` (sold/closed) |
| `creator_close` | Boolean shorthand for `creator_token_status == creator_close` |
| `dev_token_burn_ratio` | Ratio of dev's tokens that have been burned |
**Smart Money**
| Field | Description |
|-------|-------------|
| `smart_degen_count` | Number of smart money wallets holding the token |
| `renowned_count` | Number of renowned / KOL wallets holding the token |
| `bluechip_owner_percentage` | Ratio of holders that are bluechip wallets (01) |
**Social**
| Field | Description |
|-------|-------------|
| `twitter_username` | Twitter / X username (not a full URL — prepend `https://x.com/` to get the link) |
| `website` | Project website URL |
| `telegram` | Telegram URL |
| `cto_flag` | Community takeover flag (`1` = CTO has occurred) |
**Dexscreener Marketing**
| Field | Description |
|-------|-------------|
| `dexscr_ad` | Dexscreener ad placed (`1` = yes) |
| `dexscr_update_link` | Social links updated on Dexscreener (`1` = yes) |
| `dexscr_trending_bar` | Paid for Dexscreener trending bar (`1` = yes) |
| `dexscr_boost_fee` | Dexscreener boost amount paid (0 = none) |
---
## Workflow: Discover Trading Opportunities via Trending
### Step 1 — Fetch trending data
@@ -136,7 +355,16 @@ For each token, offer:
- **Deep dive**: `token info` + `token security` for full due diligence
- **Swap**: execute directly if the user is satisfied with the trending data alone
## Trenches Parameters
## `market trenches` Parameters
**Intent → `--type` mapping (always specify `--type` explicitly):**
| User intent | `--type` value |
|-------------|----------------|
| "new tokens", "just launched", "newly created", "latest tokens" | `new_creation` |
| "about to graduate", "near completion", "bonding curve almost full" | `near_completion` |
| "graduated tokens", "already on DEX", "open market tokens" | `completed` |
| No specific stage mentioned | omit `--type` (returns all three) |
| Parameter | Required | Description |
|-----------|----------|-------------|
@@ -147,7 +375,7 @@ For each token, offer:
Response fields: `data.new_creation`, `data.pump`, `data.completed` — each is an array of `RankItem` (same fields as `market trending` rank items). **Important: `data.pump` in the response corresponds to `--type near_completion` in the request. The API always returns this category under the key `pump`, not `near_completion`.**
### Response Item Key Reference
## `market trenches` Response Fields
**Basic Info**