ct-alpha

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

CT Alpha — Crypto Twitter Intelligence

CT Alpha — 加密货币Twitter情报工具

Turn X/Twitter into an actionable crypto intelligence layer. Search CT for narratives, alpha, strategies, and sentiment, then rank results using TweetRank (a PageRank-inspired credibility scoring system), extract tokens/CAs from multiple signals, detect coordinated raids, and suggest execution steps using available tools.

将X/Twitter转化为可落地的加密货币情报层。搜索CT上的叙事、Alpha信息、策略和市场情绪，然后通过TweetRank（受PageRank启发的可信度评分系统）对结果排序，从多信号中提取代币/合约地址，检测协同拉盘行为，并结合可用工具推荐执行步骤。

Overview

概述

Search: Query CT with crypto-optimized noise filters and relevancy sorting
TweetRank: Score tweets by author credibility + engagement quality + recency
Multi-Signal Token Detection: Cashtags, name-phrases, crypto URLs, contract addresses
Raid Detection: Flag tickers promoted mostly by low-credibility accounts
Trending: Detect trending tokens across multiple search queries
Watchlist: Monitor trusted CT accounts by category
Thread Hydration: Fetch full conversation threads
X Articles: Full-text extraction of long-form posts (>280 chars)
Cost Tracking: Per-session API spend tracking (~$0.005/tweet)
Dynamic Tool Discovery: Suggest follow-up actions with DeFi Llama, Backpack, Polymarket, etc.

搜索：使用针对加密货币优化的噪音过滤器和相关性排序查询CT内容
TweetRank：根据作者可信度+互动质量+时效性为推文评分
多信号代币检测：支持现金标签、名称短语、加密货币URL、合约地址提取
拉盘检测：标记主要由低可信度账号推广的代币代码
趋势检测：在多个搜索查询中发现热门代币
关注列表：按类别监控可信CT账号
线程补全：获取完整对话线程
X长文：提取长文帖子的全文内容（超过280字符）
成本追踪：按会话追踪API支出（约0.005美元/条推文）
动态工具发现：推荐使用DeFi Llama、Backpack、Polymarket等工具进行后续操作

Prerequisites

前置要求

Runtime: Bun (TypeScript runtime)
API Token: X API Bearer Token from developer.x.com (pay-per-use via xAI)
Cost: ~$0.005 per tweet read, ~$0.10 per quick search (20 tweets)

运行环境：Bun（TypeScript运行时）
API令牌：从developer.x.com获取的X API Bearer Token（通过xAI按使用付费）
成本：每条读取的推文约0.005美元，每次快速搜索（20条推文）约0.10美元

Quick Start

快速开始

Installation

安装

bash

undefined

bash

undefined

Clone the skill

克隆技能仓库

git clone https://github.com/yashhsm/skills.git cd skills/skills/ct-alpha

Run the installer (configures token, watchlist, cache)

运行安装程序（配置令牌、关注列表、缓存）

bun run install.ts

undefined

bun run install.ts

undefined

Environment Setup

环境配置

bash

undefined

bash

undefined

Set your X API Bearer Token

设置你的X API Bearer Token

export X_BEARER_TOKEN="your_token_here"

Or save to persistent env file

或保存到持久化环境变量文件

mkdir -p ~/.config/env echo 'export X_BEARER_TOKEN="your_token_here"' >> ~/.config/env/global.env source ~/.config/env/global.env

undefined

mkdir -p ~/.config/env echo 'export X_BEARER_TOKEN="your_token_here"' >> ~/.config/env/global.env source ~/.config/env/global.env

undefined

Basic Usage

基础使用

bash

undefined

bash

undefined

Search for alpha on a token

搜索某代币的Alpha信息

bun run ct-search.ts search "$SOL alpha" --quick

Detect trending tokens

检测热门代币

bun run ct-search.ts trending --window 6h --solana-only

Monitor watchlist accounts

监控关注列表账号

bun run ct-search.ts watchlist --since 24h

Read a specific tweet or article

读取特定推文或长文

bun run ct-search.ts read https://x.com/user/status/123456

Check API spending

查看API支出

bun run ct-search.ts cost

undefined

bun run ct-search.ts cost

undefined

CLI Reference

CLI参考

search — Core research command

search — 核心研究命令

bash

bun run ct-search.ts search "<query>" [flags]

Flag	Description	Default
`--quick`	20 tweets, 1hr cache, ~$0.10	Default mode
`--full`	Up to 100 tweets, 15min cache, ~$0.50	Confirm cost first
`--limit N`	Max total tweets	20 (quick), 100 (full)
`--sort <field>`	`likes` , `recency` , `relevancy`	`relevancy`
`--since <duration>`	`1h` , `6h` , `24h` , `7d`	`24h`
`--min-likes N`	Engagement filter	3 (quick)
`--from user1,user2`	Restrict to specific accounts	—
`--extract-tickers`	Show extracted tickers	—
`--extract-cas`	Show contract addresses and crypto URLs	—
`--raw`	JSON output	—

bash

bun run ct-search.ts search "<query>" [flags]

标记	描述	默认值
`--quick`	20条推文，1小时缓存，约0.10美元	默认模式
`--full`	最多100条推文，15分钟缓存，约0.50美元	需先确认成本
`--limit N`	推文总数上限	20（快速模式），100（完整模式）
`--sort <field>`	排序字段： `likes` , `recency` , `relevancy`	`relevancy`
`--since <duration>`	时间范围： `1h` , `6h` , `24h` , `7d`	`24h`
`--min-likes N`	互动量过滤器	3（快速模式）
`--from user1,user2`	限制为特定账号	—
`--extract-tickers`	显示提取的代币代码	—
`--extract-cas`	显示合约地址和加密货币URL	—
`--raw`	输出JSON格式	—

trending — Multi-signal trending detection

trending — 多信号趋势检测

bash

bun run ct-search.ts trending [flags]

Flag	Description	Default
`--window <duration>`	`1h` , `6h` , `24h`	`6h`
`--min-mentions N`	Minimum mention count	3
`--solana-only`	Solana ecosystem only	—
`--top N`	Top N results	20

bash

bun run ct-search.ts trending [flags]

标记	描述	默认值
`--window <duration>`	时间窗口： `1h` , `6h` , `24h`	`6h`
`--min-mentions N`	最低提及次数	3
`--solana-only`	仅覆盖Solana生态	—
`--top N`	显示前N个结果	20

watchlist — Monitor CT accounts

watchlist — 监控CT账号

bash

bun run ct-search.ts watchlist [flags]

Flag	Description	Default
`--category <cat>`	Filter by category	all
`--since <duration>`	Time window	`24h`

bash

bun run ct-search.ts watchlist [flags]

标记	描述	默认值
`--category <cat>`	按类别过滤	all
`--since <duration>`	时间窗口	`24h`

read — Read a specific tweet/article

read — 读取特定推文/长文

bash

bun run ct-search.ts read <tweet_url_or_id> [--thread] [--raw]

Accepts x.com URLs, twitter.com URLs, or raw tweet IDs. Articles (long-form posts) are fetched in full.

bash

bun run ct-search.ts read <tweet_url_or_id> [--thread] [--raw]

支持x.com链接、twitter.com链接或原始推文ID。长文帖子会被完整提取。

thread — Hydrate conversation thread

thread — 补全对话线程

bash

bun run ct-search.ts thread <tweet_id>

bash

bun run ct-search.ts thread <tweet_id>

cost — Track API spending

cost — 追踪API支出

bash

bun run ct-search.ts cost [--reset]

bash

bun run ct-search.ts cost [--reset]

Core Features

核心功能

TweetRank Scoring

TweetRank评分

Every tweet is scored by three factors multiplied together:

AuthorCred (0-10): Watchlist membership (+5), follower/following ratio (capped +5), verification (+1), account age (log scale, capped +2), bot penalty (-3)
EngagementQuality: Bookmarks (×3, unfakeable) > Quotes (×2.5, high effort) > Likes (×1.5) > Retweets (×1, easily botted). All log-scaled.
RecencyBoost:
```
1 / (1 + hoursAgo / 24)
```
— newer tweets score higher.

TweetRank = AuthorCred × EngagementQuality × RecencyBoost

Each tweet receives a source label:

```
[WATCHLIST]
```
— Author is on your watchlist (highest trust)
```
[HIGH-CRED]
```
— AuthorCred ≥ 5
```
[UNKNOWN]
```
— Unverified author
```
[SUSPICIOUS]
```
— Bot-like patterns detected

每条推文的得分由三个因素相乘得出：

AuthorCred（0-10）：关注列表成员（+5）、粉丝/关注者比率（上限+5）、认证标识（+1）、账号时长（对数缩放，上限+2）、机器人惩罚（-3）
互动质量：书签（×3，难以伪造）> 引用推文（×2.5，高投入）> 点赞（×1.5）> 转发（×1，易被机器人刷量）。所有指标均采用对数缩放。
时效性增益：
```
1 / (1 + hoursAgo / 24)
```
— 最新推文得分更高。

TweetRank = AuthorCred × 互动质量 × 时效性增益

每条推文会被标记来源类型：

```
[WATCHLIST]
```
— 作者在你的关注列表中（可信度最高）
```
[HIGH-CRED]
```
— AuthorCred ≥ 5
```
[UNKNOWN]
```
— 未认证作者
```
[SUSPICIOUS]
```
— 检测到类机器人模式

Multi-Signal Token Detection

多信号代币检测

Cashtag-only detection misses how tokens actually spread on CT. This extracts from four signal types:

Signal	Example	Confidence
Cashtag	`$SOL` , `$JTO`	High
Name-phrase	"pendle" + crypto context	High/Medium
Crypto URL	pump.fun, dexscreener, birdeye, jup.ag	High
Contract Address	Base58 (Solana) / 0x (Ethereum) with context	High/Low

Supported crypto URL domains: pump.fun, dexscreener.com, birdeye.so, jup.ag, raydium.io, solscan.io, etherscan.io.

仅靠现金标签检测会遗漏代币在CT上的实际传播方式。本工具从四种信号类型中提取信息：

信号类型	示例	可信度
现金标签	`$SOL` , `$JTO`	高
名称短语	"pendle" + 加密货币上下文	高/中
加密货币URL	pump.fun, dexscreener, birdeye, jup.ag	高
合约地址	Base58（Solana）/ 0x（Ethereum）+ 上下文	高/低

支持的加密货币URL域名：pump.fun, dexscreener.com, birdeye.so, jup.ag, raydium.io, solscan.io, etherscan.io.

Raid Detection

拉盘检测

Detects coordinated pump campaigns by analyzing author credibility distribution per ticker:

If >70% of authors mentioning a ticker have low credibility (AuthorCred < 3), it's flagged as a potential raid
Output includes raid score, total/low-cred author counts

通过分析每个代币代码的作者可信度分布，检测协同拉盘活动：

如果提及某代币代码的作者中超过70%为低可信度账号（AuthorCred < 3），则会被标记为潜在拉盘
输出内容包含拉盘评分、总作者数/低可信度作者数

Noise Filtering

噪音过滤

Every search auto-appends crypto noise filters:

```
-is:retweet
```
(removes retweets)
```
-"airdrop"
```
,
```
-"giveaway"
```
,
```
-"whitelist"
```
(spam removal)

-"follow and RT"

-"follow & RT"

-"free mint"

-"dm to claim"

(engagement bait)

Quick mode also adds
```
-is:reply
```

每次搜索会自动添加加密货币噪音过滤器：

```
-is:retweet
```
（移除转发）
```
-"airdrop"
```
,
```
-"giveaway"
```
,
```
-"whitelist"
```
（移除垃圾信息）

-"follow and RT"

-"follow & RT"

-"free mint"

-"dm to claim"

（移除互动诱饵）

快速模式还会添加
```
-is:reply
```

Caching

缓存

Aggressive caching prevents redundant API calls:

Cache Type	TTL	Use Case
Quick search	1 hour	Default searches
Full search	15 minutes	Deep dives
Thread	2 hours	Conversation threads
Profile	24 hours	User lookups
Watchlist	4 hours	Account monitoring

Cache is file-based (JSON) with auto-pruning on startup (24h hard limit).

激进的缓存机制可避免重复API调用：

缓存类型	过期时间	使用场景
快速搜索	1小时	默认搜索
完整搜索	15分钟	深度研究
对话线程	2小时	对话线程
账号资料	24小时	用户查询
关注列表	4小时	账号监控

缓存为基于文件的JSON格式，启动时会自动清理（硬限制24小时）。

Research Methodology

研究方法

Follow this 6-step loop for every research request:

针对每个研究请求，遵循以下6步循环：

1. Decompose

1. 拆解

Break the user's question into 1-3 targeted search queries.

Token research: search both
```
$TICKER
```
and plain name with OR
Narratives: search thematic keywords, not just token names
Strategies: include strategy/yield/APY keywords

将用户的问题分解为1-3个针对性搜索查询。

代币研究：同时搜索
```
$TICKER
```
和普通名称，用OR连接
叙事研究：搜索主题关键词，而非仅代币名称
策略研究：包含strategy/yield/APY等关键词

2. Pre-Filter

2. 预过滤

Before any API call:

Check cache (same query within TTL is free)
Noise filters are automatic
Estimate cost: Quick ~$0.10, Full ~$0.50-1.50
Narrow time window: 24h for trending, 7d for research

在调用API前：

检查缓存（相同查询在过期时间内免费）
自动应用噪音过滤器
估算成本：快速搜索约0.10美元，完整搜索约0.50-1.50美元
缩小时间窗口：趋势分析用24小时，研究用7天

3. Search

3. 搜索

Execute with

--quick

mode first (always):

bash

bun run ct-search.ts search "$TOKEN alpha" --quick --extract-tickers

始终先使用

--quick

模式执行：

bash

bun run ct-search.ts search "$TOKEN alpha" --quick --extract-tickers

4. Extract

4. 提取

Analyze TweetRank scores and trust labels. Look for extracted tickers, contract addresses, and crypto URLs.

分析TweetRank得分和信任标签。查找提取的代币代码、合约地址和加密货币URL。

5. Deep-Dive (if needed)

5. 深度研究（如有需要）

Follow high-engagement threads:
```
bun run ct-search.ts thread <id>
```
Search specific authors:
```
--from author1,author2
```
Broaden with
```
--full
```
only if quick was insufficient

跟进高互动线程：
```
bun run ct-search.ts thread <id>
```
搜索特定作者：
```
--from author1,author2
```
仅当快速搜索结果不足时，才使用
```
--full
```
模式扩大范围

6. Synthesize

6. 合成

Combine findings into actionable intelligence:

Group by theme, not by query
Highlight tickers with strong multi-signal detection
Flag raid risks
Suggest verification and execution steps

将发现整合为可落地的情报：

按主题分组，而非按查询分组
突出显示具备强多信号检测的代币
标记拉盘风险
推荐验证和执行步骤

Query String Rules

查询字符串规则

The query argument supports X API v2 operators:

Operator	Example	Description
keyword	`solana alpha`	Both words
`"exact"`	`"yield strategy"`	Exact phrase
`OR`	`$SOL OR solana`	Either term
`-`	`-airdrop`	Exclude term
`from:`	`from:username`	Tweets by user
`has:links`	`has:links`	Tweets with URLs
`lang:`	`lang:en`	Language filter
`$`	`$SOL`	Cashtag

Do NOT use these v1.1 operators (they cause 400 errors on v2 pay-per-use):

```
min_faves:N
```
,
```
min_retweets:N
```
— use
```
--min-likes
```
CLI flag instead
```
place:
```
,
```
bio:
```
,
```
sample:
```
— not available on v2

Do NOT manually include noise filters in the query — the CLI auto-appends them.

查询参数支持X API v2运算符：

运算符	示例	描述
关键词	`solana alpha`	包含两个词
`"精确短语"`	`"yield strategy"`	精确匹配短语
`OR`	`$SOL OR solana`	包含任一术语
`-`	`-airdrop`	排除该术语
`from:`	`from:username`	特定用户的推文
`has:links`	`has:links`	包含URL的推文
`lang:`	`lang:en`	语言过滤
`$`	`$SOL`	现金标签

请勿使用以下v1.1运算符（在v2按使用付费模式下会导致400错误）：

```
min_faves:N
```
,
```
min_retweets:N
```
— 请改用
```
--min-likes
```
CLI标记
```
place:
```
,
```
bio:
```
,
```
sample:
```
— v2不支持

请勿在查询中手动添加噪音过滤器 — CLI会自动追加。

Dynamic Tool Discovery

动态工具发现

After completing research, suggest execution steps using available MCP tools:

Tool Prefix	Use Case	Example
`mcp__defillama__*`	TVL, yields, fees, prices	`get_protocol_tvl("pendle")`
`mcp__backpack__*`	Exchange price, depth, trades	`backpack_get_ticker("SOL_USDC")`
`mcp__polymarket__*`	Prediction markets	`search_polymarket("solana ETF")`
`mcp__coingecko__*`	Token data, market charts	`get_id_coins("solana")`

Always frame suggestions as "verify" not "confirm" — encourage skepticism about CT alpha.

完成研究后，推荐使用可用的MCP工具执行后续步骤：

工具前缀	使用场景	示例
`mcp__defillama__*`	TVL、收益、手续费、价格	`get_protocol_tvl("pendle")`
`mcp__backpack__*`	交易所价格、深度、交易	`backpack_get_ticker("SOL_USDC")`
`mcp__polymarket__*`	预测市场	`search_polymarket("solana ETF")`
`mcp__coingecko__*`	代币数据、市场图表	`get_id_coins("solana")`

始终将建议表述为“验证”而非“确认” — 鼓励对CT Alpha信息保持怀疑态度。

Cost Protocol

成本协议

Always
--quick
first (~$0.10 for 20 tweets). Relevancy sort = best results come first.
Only increase
```
--limit
```
if 20 results are genuinely insufficient.
Display cost estimate before
```
--full
```
mode.
Cache is aggressive — same query within TTL is free.
Two-pass strategy: First search 20 results. If more depth needed on a sub-topic, do a targeted follow-up rather than re-running with higher limits.

始终先使用
--quick
模式（20条推文约0.10美元）。相关性排序=最佳结果优先。
仅当20条结果确实不足时，才增加
```
--limit
```
。
在
```
--full
```
模式前显示成本估算。
缓存机制激进 — 相同查询在过期时间内免费。
两步策略：先搜索20条结果。如果需要对某个子主题进行更深入研究，进行针对性跟进，而非重新运行更高限制的查询。

Best Practices

最佳实践

Start narrow, broaden only if needed: Specific ticker + context words first
Use
--from
for signal: Restrict to watchlist accounts for highest signal-to-noise
Use
has:links
for substance: Analytical content, not hot takes
Never present CT findings as authoritative: Always include confidence levels and risk bullets
Contract addresses are always UNVERIFIED: Verify on-chain before interacting
Two-pass research: Quick search first, then targeted deep-dives
Track spending: Use
```
cost
```
command to monitor API usage

先窄后宽：先搜索特定代币+上下文词，必要时再扩大范围
使用
--from
获取高信号：限制为关注列表账号，获得最高信噪比
使用
has:links
获取有实质内容的内容：分析性内容，而非热门观点
绝不将CT发现作为权威信息呈现：始终包含可信度水平和风险提示
合约地址始终未经验证：交互前务必在链上验证
两步研究法：先快速搜索，再进行针对性深度研究
追踪支出：使用
```
cost
```
命令监控API使用情况

Security Considerations

安全注意事项

X Bearer Token is stored in
```
~/.config/env/global.env
```
— ensure proper file permissions
Never commit tokens to version control
Contract addresses extracted from tweets are ALWAYS unverified — always verify on-chain
Watchlist data is local-only (not synced or shared)
Cache files contain tweet data — consider cleanup for sensitive research

X Bearer Token存储在
```
~/.config/env/global.env
```
中 — 确保文件权限正确
绝不要将令牌提交到版本控制
从推文中提取的合约地址始终未经验证 — 务必在链上验证
关注列表数据仅存储在本地（不同步或共享）
缓存文件包含推文数据 — 敏感研究后可考虑清理

Skill Structure

技能结构

ct-alpha/
├── SKILL.md                    # This file — agent instructions
├── ct-search.ts                # Main CLI entry point
├── setup.ts                    # Interactive setup script
├── install.ts                  # Full installer
├── lib/
│   ├── api.ts                  # X API v2 integration, pagination, caching
│   ├── extract.ts              # Multi-signal token extraction
│   ├── tweetrank.ts            # TweetRank scoring and raid detection
│   ├── format.ts               # Output formatting with trust labels
│   ├── cache.ts                # File-based caching layer
│   ├── cost.ts                 # API cost tracking
│   └── filters.ts              # Noise filtering, engagement filtering
├── resources/
│   ├── x-api.md                # X API v2 reference
│   ├── query-templates.md      # Pre-built search patterns
│   └── tool-discovery.md       # Dynamic tool suggestion map
├── data/
│   ├── known-tokens.json       # Token name → ticker mappings
│   └── watchlist.default.json  # Default watchlist categories
└── examples/
    └── basic-search.ts         # Quick start example

ct-alpha/
├── SKILL.md                    # 本文档 — 代理指令
├── ct-search.ts                # 主CLI入口
├── setup.ts                    # 交互式设置脚本
├── install.ts                  # 完整安装程序
├── lib/
│   ├── api.ts                  # X API v2集成、分页、缓存
│   ├── extract.ts              # 多信号代币提取
│   ├── tweetrank.ts            # TweetRank评分和拉盘检测
│   ├── format.ts               # 带信任标签的输出格式化
│   ├── cache.ts                # 基于文件的缓存层
│   ├── cost.ts                 # API成本追踪
│   └── filters.ts              # 噪音过滤、互动过滤
├── resources/
│   ├── x-api.md                # X API v2参考
│   ├── query-templates.md      # 预构建搜索模式
│   └── tool-discovery.md       # 动态工具推荐映射
├── data/
│   ├── known-tokens.json       # 代币名称→代码映射
│   └── watchlist.default.json  # 默认关注列表类别
└── examples/
    └── basic-search.ts         # 快速开始示例