Token导航 LogoToken导航

Agent Skill · 研究检索

trump-trade(特朗普贸易)

trump-trade 用于查找、检索和筛选相关信息,适合在 OpenClaw 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

trump-trade 是研究检索类 Agent Skill,可用宿主为 OpenClaw,通过ClawHub获取,权限为需要联网;许可证为 MIT-0,第三方扫描已通过,按安装量在研究检索类的 73,675 个 Skill 里排第 26,674。

总安装 3,525周安装 144GitHub Stars 公开资料未说明下载 1,129核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
MIT-0
版本
1.0.0

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:trump-trade(特朗普贸易)
来源仓库:https://github.com/vialiceq/trump-trade
安装命令:
openclaw skills install trump-trade
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 OpenClaw 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

ClawHubOpenClaw
openclaw skills install trump-trade

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

通过 RSS 订阅追踪特朗普官方社交媒体平台的公开言论。支持按证据强度、发布时间等维度筛选最新帖文。数据来源为 trumpstruth.org 提供的结构化 feed 接口。仅反映单一信源视角,建议对比其他媒体交叉验证关键主张。无法获取私密通信或非公开场合发言记录。

trump-trade 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 OpenClaw;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 openclaw skills install trump-trade;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它需要需要联网,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

trump-trade 怎么安装?

在终端执行 openclaw skills install trump-trade。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

trump-trade 支持哪些 AI 工具?

公开资料标注支持 OpenClaw。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

trump-trade 需要什么权限?

权限级别是需要联网。该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

trump-trade 是免费的吗?

来源标注的许可证是 MIT-0。本站不代收任何费用,也不托管安装包;能否商用、要不要署名,以许可证原文为准。

trump-trade 用的人多吗?

公开数据是累计安装 3,525 次,近一周安装 144 次,下载 1,129 次。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

trump-trade 安全吗?

第三方扫描结果为通过,但这只覆盖已知的静态检查项,不能替代自己读一遍源码和安装命令。

值得选的地方和要先想清楚的地方

值得选的地方

  • 专注 OpenClaw 一种宿主,触发方式和使用习惯不用来回切换。
  • 许可证明确为 MIT-0,商用和分发边界好判断。
  • 第三方扫描已通过,常见的高风险行为没有命中。

先想清楚的地方

  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 OpenClaw,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。
  • 需要许可证清楚、并且有第三方扫描结果的团队项目。

不太适合

  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
trump-trade

特朗普贸易 · 本页

3,525—需要联网MIT-0通过

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

name
trump-trade
description
|

trump-trade

Use only for education and research. - No profit guarantees. - No personalized buy/sell instructions. - If the user asks for a trade decision, switch to scenario / risk discussion.

Mode detection

1) Evidence mode (must show post evidence)

Use Evidence mode when the user explicitly asks what Trump posted, his latest post, the original text, or RSS proof.

Hard rules:

  1. Show 2-5 most relevant RSS items first.
  2. Each evidence item must include:

- published time - title - link - a cleaned short excerpt from description/content (keep it short, strip HTML, avoid rewriting the meaning)

  1. Immediately under each evidence item, include “Financial analysis.”

- Explain the likely market path in educational terms: risk appetite, inflation expectations, rates, USD, oil/energy, gold, etc. - Do not claim certainty.

  1. End Evidence mode with:

- risk warning - observation checklist - directional scenario sketch (Base / Bull / Bear)

2) Latest mode

Use when the user says latest / recent / just now / what did he post most recently / latest theme and no explicit time window is given.

3) Time mode

Use when the user gives a date or range (for example: 2026-04-01, 2026-04-01 09:00, from/to, or “past N days”).

  • Parse the time window.
  • Keep items whose pubDate falls inside that window.
  • If timezone is omitted, default to Asia/Hong_Kong.

4) Trump-style opinion mode

Use when the user asks “What does Trump think about oil / USD / gold / stocks / bonds / rates / a sector?” but does not ask for post evidence.

  • Do not show raw RSS evidence.
  • Reply in a Trump-style voice, but keep it educational.
  • Include:

- risk warning - observation checklist - Base / Bull / Bear scenarios

5) K-line / chart / “dwang” style requests

If the user mentions chart-artist / draw the candles / Trump genius / similar wording:

  • If they also say latest, use Latest mode.
  • If they provide time, use Time mode.
  • Add a technical / candle-style qualitative sketch:

- possible pattern - what would confirm it - clearly state this is educational and not based on live chart data

6) Watch mode (incremental push)

Use when the user or system asks to monitor every N minutes / every 5 minutes / watch continuously.

Behavior:

  • Fetch the RSS feed every run.
  • Identify each item by truth:originalId if present, otherwise by the last part of the link.
  • Track lastSeenId in conversation context.
  • If there are new items after lastSeenId:

- output only the 2-5 new items - use the Evidence template - add “Financial analysis” under each item - say how many new items were found

  • If there are no new items:

- output a short “No updates.”

Note: actual 5-minute triggering requires external scheduling. This skill handles incremental detection and output format.


Asset / theme mapping

Map user keywords to analysis angles:

  • oil / crude / energy -> US energy stocks, supply-demand, inflation path
  • USD / DXY / FX -> rates, risk appetite, import costs
  • gold / metals -> safe haven demand, real rates
  • bonds / yields / 10Y / YTM -> rate expectations, term premium
  • stocks / S&P / Nasdaq -> risk appetite, earnings, macro factors
  • inflation / CPI -> inflation expectations and rates
  • jobs / unemployment -> growth momentum and policy expectations
  • sector keywords -> sector sentiment and factor exposure

RSS retrieval and distillation

Retrieval

  • Fetch: https://www.trumpstruth.org/feed
  • If the fetch is not easy XML, parse <item>...</item> from raw text.
  • Extract: pubDate, title, link, description/content.

Pick 2-5 items

  • Score items by overlap between the user question / mapped asset theme and item title + description.
  • If Evidence mode has weak overlap, fall back to the newest items.

Ongoing distillation

After each Evidence / Latest / Time run, keep a short distilled summary:

  • main narrative
  • possible market variables
  • observation checklist

Use that distilled summary later for Trump-style opinion mode, but do not claim you verified raw details that were not shown.


Output format

A) Evidence mode

Start with:

  • “Below are public post evidence items (education only), followed by market analysis and observations.”

Then list 2-5 items:

  • [Evidence #1]

- Published: - Title: - Link: - Excerpt: - Financial analysis:

Then include:

  1. Distilled takeaways (3-6 bullets)
  2. Observation / directional scenario notes (no buy/sell)
  3. Risk warning

B) Trump-style opinion mode

  1. Start in Trump-style voice.
  2. Give educational market commentary.
  3. Add observation checklist.
  4. Add Base / Bull / Bear scenarios.
  5. Add risk warning.

Compliance rules

  • Never output guaranteed returns, certain direction calls, or direct buy/sell instructions.
  • If the user wants a trade decision, convert it to risk / scenario discussion.
  • Do not invent raw post text. Any post detail must come from Evidence mode excerpts or be clearly marked as a summary / inference.

平台分布

OpenClaw

74.85%
按下载量换算845

安全审计

VirusTotal

通过

ClawScan

通过

Static analysis

通过

权限和风险

需要联网

该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills