← Back to Skills Marketplace
lj22503

Fund Signal Monitor

by lj22503 · GitHub ↗ · v1.1.0 · MIT-0
cross-platform ⚠ suspicious
128
Downloads
0
Stars
1
Active Installs
2
Versions
Install in OpenClaw
/install fund-signal-monitor
Description
[何时使用]当用户需要监控基金信号时;当基金季报/年报发布时;当基金经理变更时;当基金规模大幅变化时;当基金业绩异常波动时;当检测到风格漂移时。与 fund-analyzer-pro 配合使用:fund-analyzer-pro 负责主动分析 + 信号监控,本 skill 负责被动监控 + 自动推送(实际实现已集...
README (SKILL.md)

Fund-Signal-Monitor 基金信号监控 🚨

被动监控基金重要信号,自动推送提醒

版本:v1.1.0
最后更新:2026-04-28


📋 功能描述

与 fund-analyzer-pro 的关系

维度 fund-analyzer-pro fund-signal-monitor
定位 主动分析工具 被动监控工具
触发 用户主动询问 自动监控 + 推送
职责 八大模块分析 五大信号监控
输出 完整分析报告 简短提醒信号

协作方式

用户监控列表 → fund-signal-monitor 监控
                    ↓ 检测到信号
                    ↓ 推送提醒
                    ↓
              用户点击详情
                    ↓
              fund-analyzer-pro 深度分析

🚨 五大信号

信号 1:季报/年报发布 📊

触发条件

  • 基金公司披露季报/年报
  • 距离上次报告>3 个月

监控频率:每日 15:30(收盘后)

报告内容

# 基金信号:季报发布

**涉及基金**:XXX (000001)

**信号内容**:
- 2026 年 Q1 季报已发布
- 规模:XX 亿(环比 +X%)
- 重仓股变化:新增 XX,减持 XX

**影响评估**:
- 对业绩影响:[正面/负面/中性]
- 建议操作:[关注/持有/调整]

**一键分析**:[调用 fund-analyzer-pro 深度分析]

信号 2:基金经理变更 👤

触发条件

  • 经理离职公告
  • 新任经理公告

监控频率:实时(公告即推送)

报告内容

# 基金信号:经理变更

**涉及基金**:XXX (000001)

**信号内容**:
- 离任经理:张坤(任职 X 年,回报 +X%)
- 新任经理:XXX(从业 X 年,代表作 XXX)

**影响评估**:
- 对业绩影响:待观察
- 对风险影响:[高/中/低]
- 建议操作:持有观察

**一键分析**:[调用 fund-analyzer-pro 分析新任经理]

信号 3:基金规模大幅变化 📈

触发条件

  • 规模环比增长>50%
  • 规模环比下降>30%

监控频率:每周一次(周一 10:00)

报告内容

# 基金信号:规模变化

**涉及基金**:XXX (000001)

**信号内容**:
- 当前规模:XX 亿
- 环比变化:+X%(大幅增长)
- 原因:[业绩好/营销/分红]

**影响评估**:
- 规模过大可能影响灵活性
- 建议操作:[关注/持有/调整]

**一键分析**:[调用 fund-analyzer-pro 分析规模影响]

信号 4:业绩异常波动 📉

触发条件

  • 单日涨跌>5%
  • 连续 3 日涨跌>10%

监控频率:实时(盘中监控)

报告内容

# 基金信号:业绩异常

**涉及基金**:XXX (000001)

**信号内容**:
- 今日涨跌:+X%(异常波动)
- 可能原因:[重仓股大涨/行业利好]

**影响评估**:
- 短期波动 vs 长期趋势
- 建议操作:[持有/加仓/减仓]

**一键分析**:[调用 fund-analyzer-pro 分析波动原因]

信号 5:风格漂移 🎭

触发条件

  • 持仓风格与宣称不符
  • 行业配置大幅变化(>20%)

监控频率:每次季报发布后

报告内容

# 基金信号:风格漂移

**涉及基金**:XXX (000001)

**信号内容**:
- 宣称风格:价值成长
- 实际持仓:周期股占比 X%(大幅偏离)
- 行业变化:消费 -20%,周期 +20%

**影响评估**:
- 风格漂移可能增加风险
- 建议操作:[关注/调整]

**一键分析**:[调用 fund-analyzer-pro 分析风格一致性]

🔄 使用流程

流程 1:设置监控列表

用户添加监控基金
  ↓
保存到用户配置
  ↓
启动定时任务

配置示例

# 用户监控配置
monitor_list:
  - code: 000001
    name: 易方达蓝筹精选
    signals: [季报,经理,规模,波动,风格]
  - code: 163406
    name: 兴全合润
    signals: [季报,经理,波动]

schedule:
  - signal: 季报
    time: 每日 15:30
  - signal: 经理
    time: 实时
  - signal: 规模
    time: 每周一 10:00
  - signal: 波动
    time: 实时
  - signal: 风格
    time: 季报发布后

流程 2:监控 + 推送

定时任务触发
  ↓
调用数据源(天天基金/且慢 MCP)
  ↓
检测信号(阈值判断)
  ↓
检测到信号 → 生成报告 → 推送给用户
  ↓
用户点击详情 → 调用 fund-analyzer-pro 深度分析

流程 3:与 fund-analyzer-pro 联动

用户收到信号提醒
  ↓
点击"一键分析"
  ↓
调用 fund-analyzer-pro
  ↓
输出完整分析报告
  ↓
用户决策

示例

用户:收到"经理变更"信号

→ 点击"一键分析"
→ 调用 fund-analyzer-pro 模块 5(基金经理分析)
→ 输出新任经理深度分析
→ 用户决定是否继续持有

📝 输出示例

示例 1:经理变更推送

🚨 基金信号提醒

**信号类型**:基金经理变更

**涉及基金**:易方达蓝筹精选 (000001)

**信号内容**:
- 张坤离任,新任经理:XXX
- 离任日期:2026-04-15
- 新任日期:2026-04-16

**影响评估**:
- 对业绩影响:待观察
- 对风险影响:中性
- 建议操作:持有观察

**跟踪要点**:
- [ ] 新任经理历史业绩
- [ ] 持仓风格是否变化
- [ ] 下季度报告

---
[一键分析] 调用 fund-analyzer-pro 深度分析
[忽略] 暂时不处理

示例 2:季报发布推送

📊 基金信号提醒

**信号类型**:季报发布

**涉及基金**:兴全合润 (163406)

**信号内容**:
- 2026 年 Q1 季报已发布
- 规模:150 亿(环比 +5%)
- 重仓股变化:新增 XX,减持 XX

**影响评估**:
- 对业绩影响:正面
- 建议操作:继续持有

---
[一键分析] 调用 fund-analyzer-pro 深度分析
[忽略] 暂时不处理

🔧 技术实现

⚠️ 注意:实际信号检测引擎已集成到 fund-analyzer-pro 的 signal_checker.py 中。 本 skill 保留为文档参考,实际使用请调用 fund-analyzer-pro 的信号监控功能。

信号检测引擎(fund-analyzer-pro)

脚本位置skills/fund-analyzer-pro/scripts/signal_checker.py

运行方式

# 检查所有基金
python signal_checker.py

# 检查指定基金
python signal_checker.py 005827

# 模拟运行
python signal_checker.py --dry-run

# 每日摘要
python signal_checker.py --summary

监控列表配置

配置文件skills/fund-analyzer-pro/watchlist.json

CLI 管理

# 列出监控
python signal_checker.py --list

# 添加监控
python signal_checker.py --add 005827 易方达蓝筹精选

# 移除监控
python signal_checker.py --remove 005827

定时任务配置(heartbeat 触发)

# 在 HEARTBEAT.md 中添加:
# - 交易日 15:30:执行 signal_checker.py(季报/规模/波动检查)
# - 每周一 10:00:执行 signal_checker.py --summary(每日摘要)

数据源

信号 数据源
季报/年报 天天基金 API
经理变更 天天基金 API + 且慢 MCP
规模变化 天天基金 API
业绩波动 新浪 API(实时)
风格漂移 且慢 MCP(持仓明细)

与 fund-analyzer-pro 联动

# 检测到信号后自动联动 fund-analyzer-pro
# 用户回复 "分析 005827" → 调用 fund-analyzer-pro 深度分析

⚠️ 注意事项

1. 信号频率控制

避免过度推送

  • 同一基金同一信号,24 小时内只推送一次
  • 非交易时间不推送(除经理变更外)
  • 用户可自定义推送频率

2. 数据准确性

数据源验证

  • 季报数据:以基金公司官网为准
  • 经理变更:以公告为准
  • 规模数据:可能存在 1-2 天延迟

3. 免责声明

每份报告必须包含

本文内容仅供参考,不构成任何投资建议。市场有风险,投资需谨慎。


🔗 相关资源

  • fund-analyzer-pro - 基金分析主 skill(主动分析)
  • ttfund-skills - 天天基金 API
  • qieman-mcp - 且慢 MCP

📊 版本历史

版本 日期 更新内容
v1.1.0 2026-04-28 指向 fund-analyzer-pro signal_checker.py 实现
v1.0.0 2026-04-16 初始版本(五大信号 + fund-analyzer-pro 联动)

版本:v1.1.0
类型:通用🟡
数据源:天天基金 API + 且慢 MCP
联动:fund-analyzer-pro(signal_checker.py)

Usage Guidance
What to check before installing: 1) Confirm fund-analyzer-pro is installed and inspect skills/fund-analyzer-pro/scripts/signal_checker.py — this skill mostly documents that script rather than implementing detection itself. 2) Ask the author or inspect repository for where API keys and push credentials are stored/configured. The manifest lists no required env vars, yet the docs reference APIs and Feishu/WeChat pushes — don't provide unrelated secrets until you confirm how/where they're used. 3) Verify which account/endpoints will receive push notifications (Feishu/WeChat webhooks, your personal token, or a shared service). Prefer service-specific scoped tokens and short-lived credentials. 4) Test in dry-run mode (the docs mention --dry-run) and run locally before enabling any automatic scheduling to ensure it only accesses the expected APIs and file paths. 5) Review persistence locations (where monitor lists are saved and heartbeat entries are written) and ensure they don't overwrite unrelated configs. If you need stronger assurance, ask for an explicit requires.env listing and a minimal example of push configuration and permissions. 6) If you want higher assurance, request an explicit manifest update adding required env vars and push endpoints or ask the maintainer to provide an implementation that uses platform-approved secret storage.
Capability Analysis
Type: OpenClaw Skill Name: fund-signal-monitor Version: 1.1.0 The fund-signal-monitor skill bundle is a legitimate financial monitoring tool designed to track fund manager changes, quarterly reports, and performance volatility. The documentation in SKILL.md and the assessment files describe a clear, non-malicious architecture that integrates with fund-analyzer-pro and utilizes public financial APIs (Tiantian Fund, Sina). No evidence of data exfiltration, malicious code execution, or harmful prompt injection was found; the requested capabilities (scheduling and script execution) are consistent with the tool's stated purpose of automated monitoring.
Capability Tags
requires-sensitive-credentials
Capability Assessment
Purpose & Capability
The skill is an instruction-only passive fund signal monitor that delegates actual signal detection to fund-analyzer-pro's signal_checker.py. The stated purpose (monitoring five fund signals and pushing alerts) aligns with the SKILL.md content and related_skills. No unrelated capabilities (e.g., cloud admin, SSH, or unrelated APIs) are requested.
Instruction Scope
SKILL.md stays within monitoring and push workflows and documents data sources (天天基金 API, 且慢 MCP, 新浪 API) and CLI usage for fund-analyzer-pro. It instructs saving user monitor lists, starting scheduled tasks (heartbeat), and sending push notifications (observations mention 飞书/微信). However the instructions are high-level and assume existence of fund-analyzer-pro and push integrations; they could cause an agent to request or use credentials at runtime because the skill does not declare how push/auth is configured.
Install Mechanism
This is instruction-only with no install spec and no files that write/execute code as part of installation. That minimizes install-time risk.
Credentials
SKILL.md references APIs and push channels that normally require credentials, and the ASSESSMENT doc mentions 'API Key via environment variables', yet the skill's metadata lists no required environment variables or primary credential. Push channels (Feishu/WeChat) and data sources are not declared as needing tokens. This mismatch means the agent or skill may ask for or expect secrets at runtime without them being declared in the skill manifest—unexpected credential requests are the main concern.
Persistence & Privilege
always is false and autonomous invocation is allowed (platform default). The instructions tell the agent to save monitor lists and configure scheduled tasks (heartbeat entries). That is plausible for this skill but does imply persistence (saved configs and scheduled jobs) and push capability — the user should verify where configuration is stored and which account/endpoint will be used for notifications.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install fund-signal-monitor
  3. After installation, invoke the skill by name or use /fund-signal-monitor
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.1.0
指向 fund-analyzer-pro signal_checker.py 实现
v1.0.0
基金信号监控 v1.0.0 - 被动监控五大信号(季报/经理/规模/波动/风格),与 fund-analyzer-pro 联动,skill-evolve 评分 9.2/10
Metadata
Slug fund-signal-monitor
Version 1.1.0
License MIT-0
All-time Installs 1
Active Installs 1
Total Versions 2
Frequently Asked Questions

What is Fund Signal Monitor?

[何时使用]当用户需要监控基金信号时;当基金季报/年报发布时;当基金经理变更时;当基金规模大幅变化时;当基金业绩异常波动时;当检测到风格漂移时。与 fund-analyzer-pro 配合使用:fund-analyzer-pro 负责主动分析 + 信号监控,本 skill 负责被动监控 + 自动推送(实际实现已集... It is an AI Agent Skill for Claude Code / OpenClaw, with 128 downloads so far.

How do I install Fund Signal Monitor?

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

Is Fund Signal Monitor free?

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

Which platforms does Fund Signal Monitor support?

Fund Signal Monitor is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Fund Signal Monitor?

It is built and maintained by lj22503 (@lj22503); the current version is v1.1.0.

💬 Comments