← Back to Skills Marketplace
rayniubi

FollowinMCP

by ray · GitHub ↗ · v1.0.0 · MIT-0
darwinlinuxwin32 ✓ Security Clean
77
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install followin-mcp
Description
Followin MCP 相关子 Skill 的索引与入口。在需要路由到情报/舆情/策略/宏观/美股等能力时,先读本文件再打开 references 下对应 *.SKILL.md。
README (SKILL.md)

Followin MCP — 子 Skill 索引

本目录的 references/ 下包含 13 个独立 Skill 文件(*.SKILL.md),每个文件自带 YAML name / descriptionmetadata,以及 「本 Skill 与 MCP 工具映射」(含 followin-mcp / premium-mcp SSE 配置片段)。
实现具体任务时:根据用户意图选中下表中的文件,完整阅读该文件正文,再按其中的工具名与路由规则调用 MCP。

打包与目录结构说明见 references/README.md


按主题分组

情报与舆情(加密)

文件 Skill 名 用途概要
references/01-followin-intel-center.SKILL.md followin-intel-center 代币解锁、宏观数据、上币下架、项目事件、量价异动、资金费率等「情报中心」聚合频道。
references/02-breaking-news.SKILL.md breaking-news 针对单条加密新闻/事件,分析受影响标的与多空倾向;正文以分析流程为主,无固定 MCP 工具
references/03-trending-news-topics.SKILL.md trending-news-topics 「今天发生什么、什么最热」类热点舆情;关注事件与话题,非单币深度。
references/04-crypto-daily-brief.SKILL.md crypto-daily-brief 加密市场日报/早报(触发词如日报、加密日报);宏观/美股早报不归本 Skill。
references/05-token-buzz-views.SKILL.md token-buzz-views 指定具体代币后的新闻、深度、KOL、社群观点聚合。

交易与社群信号

文件 Skill 名 用途概要
references/06-trading-strategy-signal.SKILL.md trading-strategy-signal 顶级交易员实盘、KOL 喊单等交叉验证;模糊「市场/KOL 怎么看」及所有 KOL 喊单类查询默认归本 Skill;不含资金费率
references/07-tg-channel-intel.SKILL.md tg-channel-intel Telegram 频道话题、共识与分歧;可按偏好筛选频道。

宏观看盘与早报

文件 Skill 名 用途概要
references/08-btc-macro-dashboard.SKILL.md btc-macro-dashboard BTC 宏观环境 0–100 评分与分层解读(如「BTC 宏观怎么样」)。
references/09-gold-macro-dashboard.SKILL.md gold-macro-dashboard 黄金宏观环境 0–100 评分与分层解读。
references/10-macro-morning-brief.SKILL.md macro-morning-brief 宏观/美股维度晨间简报(触发词含宏观日报、美股早报、morning brief 等);纯「加密日报」走 crypto-daily-brief

美股与宏观分析

文件 Skill 名 用途概要
references/11-us-stock-earnings-report.SKILL.md us-stock-earnings-report 单股财报三维分析(须指定代码或公司名);泛问「今天有哪些财报」走情报中心。
references/12-macro-analyzer.SKILL.md macro-analyzer 宏观指标 影响/解读(须同时有指标名 + 影响类意图);纯数据查询走情报中心,综合晨报走 macro-morning-brief,BTC/黄金盘面走对应 dashboard。
references/13-us-stock-divergence-scan.SKILL.md us-stock-divergence-scan 价格、媒体与内部人交易等背离/静默异动扫描;泛问异常走热点舆情,个股异动规则见各 Skill 边界说明。

快速路由(与其它 Skill 的边界)

  • 「加密日报」04-crypto-daily-brief「宏观/美股早报、morning brief」10-macro-morning-brief
  • 「今天有什么热点/消息」03-trending-news-topics「某代币新闻/KOL 观点」05-token-buzz-views(须点名代币)。
  • 「KOL 喊单 / 交易员实盘」06-trading-strategy-signal资金费率/解锁/上币等频道情报01-followin-intel-center
  • 「BTC 或黄金宏观打分」08 / 09「某宏观指标对市场的影响」12-macro-analyzer
  • 各文件 description 与正文中的 不得路由到其它 Skill 说明具有最高优先级;冲突时以目标 Skill 全文为准。

使用方式(给 Agent)

  1. 根据用户问题在上表中定位 一个 主 Skill 文件。
  2. 打开对应 references/*.SKILL.md,遵循其 MCP 配置工具映射表
  3. 需要并列能力时(例如情报中心 + 单币舆情),分别打开多个文件,注意各 Skill 声明的互斥路由。
Usage Guidance
This package is an index that routes queries to a set of Followin MCP-backed skills. Before installing or enabling it: 1) Verify and trust the external endpoints (mcp.followin.io / premium-mcp.followin.io) because the skill's runtime expects you to configure API keys for those endpoints in your host configuration. 2) The skill doesn’t itself request env vars, but you will need to supply MCP API keys in your agent/host config — never paste those keys into untrusted places. 3) Several referenced skills instruct using web searches and (optionally) local tools or shell processing for large payloads; if your agent runtime has shell access, consider the security posture of that environment. 4) Source/homepage is missing — if provenance matters for you, ask the publisher for a verified homepage or code repository and verify the owner before supplying credentials. 5) If you want further assurance, request a proof-of-origin (signed repo, official Followin docs) or run the referenced skills in a sandboxed agent instance first.
Capability Analysis
Type: OpenClaw Skill Name: followin-mcp Version: 1.0.0 The skill bundle is a comprehensive set of financial analysis tools for cryptocurrency and US stock markets, designed to interface with external Model Context Protocol (MCP) servers at followin.io. The instructions (SKILL.md files) provide detailed guidance for the AI agent on intent routing, data filtering, and structured reporting, including specific constraints to remove investment advice and prioritize high-risk signals. No evidence of data exfiltration, malicious code execution, or harmful prompt injection was found; the bundle's behavior is entirely consistent with its stated purpose of providing market intelligence.
Capability Tags
cryptocan-make-purchasesrequires-sensitive-credentials
Capability Assessment
Purpose & Capability
The top-level file's name/description (an index and router for Followin MCP sub-skills) matches what is present: 13 referenced SKILL.md files that document MCP tool mappings and routing rules. The external SSE endpoints and finance/mcp tools referenced are coherent with the stated financial/intel use cases.
Instruction Scope
All runtime instructions are documentation for selecting and following the referenced Skill files and for calling MCP tools (SSE endpoints like mcp.followin.io / premium-mcp.followin.io). Several referenced Skill files instruct the agent to call MCP tools, perform web lookups as a fallback, and (in a few places) use agent tooling or 'bash' to post-process large returned payloads. This is expected for data-heavy financial skills but means the agent will rely on external MCP endpoints and may invoke local tooling if available.
Install Mechanism
Instruction-only package with no install spec and no code files. Nothing is written to disk by the skill itself; there is no package download or third‑party install to review.
Credentials
The skill declares no required environment variables, but many referenced files explicitly show how to configure MCP SSE endpoints using API keys (placeholders like YOUR_API_KEY_HERE). That is expected for a connector to Followin's MCP servers, but the skill itself does not declare or request any particular secret names — the user or host must provide MCP credentials in the host configuration. No unrelated credentials (AWS, GitHub tokens, etc.) are requested.
Persistence & Privilege
No always:true, no install scripts, and the skill does not request to modify other skills or system-wide settings. It is user-invocable and allows autonomous invocation by the agent (platform default), which is normal for skills that route tasks.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install followin-mcp
  3. After installation, invoke the skill by name or use /followin-mcp
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
- Initial release of followin-mcp as an index and entry point for related MCP sub-skills. - Provides a categorized table of 13 sub-skills with file links, names, and summary of use cases. - Includes quick routing rules to help select the correct Skill based on user intent. - Offers guidance on usage boundaries and priority rules when routing between Skills. - Detailed usage instructions for agents on how to locate and use sub-skills.
Metadata
Slug followin-mcp
Version 1.0.0
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 1
Frequently Asked Questions

What is FollowinMCP?

Followin MCP 相关子 Skill 的索引与入口。在需要路由到情报/舆情/策略/宏观/美股等能力时,先读本文件再打开 references 下对应 *.SKILL.md。 It is an AI Agent Skill for Claude Code / OpenClaw, with 77 downloads so far.

How do I install FollowinMCP?

Run "/install followin-mcp" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is FollowinMCP free?

Yes, FollowinMCP is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does FollowinMCP support?

FollowinMCP is cross-platform and runs anywhere OpenClaw / Claude Code is available (darwin, linux, win32).

Who created FollowinMCP?

It is built and maintained by ray (@rayniubi); the current version is v1.0.0.

💬 Comments