← Back to Skills Marketplace
cr330326

fund-screening

by Allen Chen · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ✓ Security Clean
92
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install fund-screening
Description
基金筛选与定投实战技能。使用五维筛选体系(业绩/经理/风格/持仓/机构)从天天基金网和晨星网筛选优质基金,构建投资组合并执行定投策略。触发场景:用户提到"基金筛选"、"基金定投"、"选基"、"筛选基金"、"基金组合"、"定投策略"、"fund screening"、"fund DCA",或要求推荐/分析/对比基金...
README (SKILL.md)

基金筛选与定投实战

概述

基于七年实战经验沉淀的五维基金筛选体系,覆盖完整闭环:初筛 → 精筛 → 深度分析 → 回测验证 → 组合配置 → 定投执行 → 定期检视

完整策略文档:实战/基金/基金筛选投资策略.md

核心工作流

用户请求 → 判断任务类型
  ├─ A. 全流程筛选(从零开始选基)
  ├─ B. 单只基金深度分析
  ├─ C. 组合配置 / 调仓建议
  ├─ D. 定投执行与检视
  └─ E. 读取/更新已有筛选数据

A. 全流程筛选

依次执行以下 4 步,每步产出明确:

第 1 步:天天基金网初筛

  • 搜索天天基金网,筛选混合型/股票型基金
  • 条件:近 3 年、2 年、1 年、6 月、3 月涨幅同时进同类前 50
  • 产出:初筛名单(~30-50 只)

第 2 步:晨星网精筛

  • 搜索晨星网,交叉验证初筛名单
  • 条件:三年评级 ≥ 三星,五年评级 ≥ 五星,夏普比率前 20~40,晨星风险系数低于同类平均
  • 产出:候选池(~10 只)

第 3 步:五维深度分析(详见 references/five-dimensions.md

  • ① 过往业绩稳定性
  • ② 基金经理评分、任期、最大回撤
  • ③ 投资风格与晨星风格箱
  • ④ 前 10 大持仓分析
  • ⑤ 机构持有占比与资金流向
  • 产出:每只基金的评分卡

第 4 步:定投回测验证

  • 分别在牛市和熊市场景下回测
  • 牛市年化 ≥ 30%,熊市年化 ≥ 10% 为达标
  • 产出:最终精选 4 只入组合

B. 单只基金深度分析

对指定基金代码执行五维分析,输出结构化评分卡:

基金名称:XXX(代码)
─────────────────────
① 业绩:近3年/2年/1年/6月/3月 同类排名 → ⭐⭐⭐⭐⭐
② 经理:评分 X/10 | 任期 X年 | 最大回撤 X% → ⭐⭐⭐⭐
③ 风格:晨星风格箱 [X盘X] | 规模 X亿 → ⭐⭐⭐⭐
④ 持仓:前10持仓集中度 X% | 行业分布 → ⭐⭐⭐
⑤ 机构:机构占比 X% | 近1年净申购/赎回 → ⭐⭐⭐⭐
─────────────────────
综合评级:★★★★☆ | 建议:适合/不适合定投

C. 组合配置

参考晨星风格箱,从候选池中配置:

角色 数量 风格
核心 2 只 中盘成长
稳定器 1 只 大盘平衡
补充 1 只 小盘平衡(量化)

按市场环境调整:

  • 牛市 → 全配股票型
  • 震荡市 → 混合型为主 + 债券型
  • 熊市 → 增加债券型、货币型

D. 定投执行与检视

定投纪律:

  • 不止损、要止盈(20%)、不断供
  • 大跌大买、小跌小买、不跌正常定投
  • 买入时机:工作日 14:00~15:00

定期检视(每 2-3 个月):

  1. 同类排名是否仍前列
  2. 基金经理是否更换
  3. 投资风格是否漂移
  4. 规模是否异常变动
  5. 机构持有比例变化
  6. 净赎回是否持续
  7. 市场环境是否需调整配比

E. 读取已有数据

工作区已有筛选记录:

  • 实战/基金/结果/天天基金筛选20201024.xlsx — 天天基金网筛选结果
  • 实战/基金/结果/晨星网基金筛选.xlsx — 晨星网筛选结果
  • 实战/基金/结果/基金定投记录表.xlsx — 定投跟踪记录

读取 XLSX 使用 uv run --with openpyxl python3uv run --with pandas python3

数据源

平台 用途 关键功能
天天基金网 fund.eastmoney.com 初筛、排名、规模、申赎数据 基金排行、基金对比
晨星网 cn.morningstar.com 精筛、评级、风格箱、夏普比率 基金筛选器、风险评价
好买基金网 howbuy.com 基金经理评级 经理评分、任期查询

使用 tavily-search / tavily-extract / agent-browser 技能获取实时数据。

输出规范

  • 所有输出使用中文
  • 涉及具体基金推荐时必须声明:「仅供参考,不构成投资建议」
  • 筛选结果保存为 Markdown 表格或 XLSX 到 实战/基金/结果/ 目录
  • 文件命名遵循:{平台}筛选{YYYYMMDD}.xlsx基金筛选结果{YYYYMMDD}.md

参考资料

  • 五维分析详解: references/five-dimensions.md — 每个维度的具体评判标准与阈值
  • 完整策略原文: 实战/基金/基金筛选投资策略.md — 从学习笔记提炼的完整策略文档
Usage Guidance
This skill appears to do exactly what it says: screen funds using public data sources, analyze them, run backtests, and save results to the listed workspace files. Before installing or enabling it, consider: (1) Are you comfortable the agent will read/write the specific workspace files listed (实战/基金/结果/*.xlsx)? If not, move or protect those files. (2) The SKILL.md expects the agent to run Python with openpyxl/pandas via an 'uv' runner and to call other skills for web scraping — ensure those runtimes/skills are available and trusted. (3) The skill fetches data from public websites (eastmoney, morningstar, howbuy); scraping those sites may have legal/ToS implications and could change if site structure changes. (4) The skill includes clear disclaimers about not being investment advice, but you should still treat recommendations as informational only. If you want to reduce risk, disable autonomous invocation for the agent or limit its file access and network capabilities.
Capability Analysis
Type: OpenClaw Skill Name: fund-screening Version: 1.0.0 The skill bundle is a legitimate tool for fund screening and investment analysis using data from Eastmoney and Morningstar. It utilizes standard agent capabilities like web searching (Tavily), browser automation, and Python execution (via uv run) for processing Excel files, all of which are directly aligned with its stated financial analysis purpose. No indicators of data exfiltration, malicious execution, or prompt injection were found in SKILL.md or the reference documents.
Capability Assessment
Purpose & Capability
The name/description (fund screening, DCA execution, portfolio construction) matches the runtime instructions: web queries to fund sites, multi-step screening, local XLSX read/write and backtests. Minor mismatch: the SKILL.md tells the agent to run commands like `uv run --with openpyxl python3` and to rely on other skills (tavily-search/tavily-extract/agent-browser) but the skill metadata does not declare required binaries or explicit dependencies for running Python/openpyxl/pandas or the 'uv' runner. This is a usability/declared-dependency gap rather than evidence of malicious intent.
Instruction Scope
Instructions stay within the stated scope (initial screening, precision filtering, five-dimension analysis, backtests, saving results). The skill explicitly reads and writes workspace files (listed XLSX paths) — which is expected for a screening skill but is something the user should be aware of. It instructs the agent to fetch real-time data from public fund websites via other skills, which is coherent but increases runtime dependencies and surface area.
Install Mechanism
Instruction-only skill with no install spec and no code files included. This lowers installation risk because nothing is downloaded or executed during install.
Credentials
The skill requests no credentials, no environment variables, and no config paths. Its external data access is limited to public fund websites (eastmoney, morningstar, howbuy) via other agent skills — proportional for the stated purpose.
Persistence & Privilege
always is false and the skill is user-invocable (normal). It does not request permanent presence or system-wide configuration changes.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install fund-screening
  3. After installation, invoke the skill by name or use /fund-screening
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
- 首次发布基金筛选与定投实战技能,基于五维筛选体系实现全流程选基与组合定投。 - 支持天天基金网和晨星网多维度筛选、质量验证,并输出结构化评分卡。 - 覆盖从初筛、精筛、深度分析到回测验证、组合配置及定投执行的完整闭环流程。 - 明确输出规范,简化定投记录与组合调整,支持读取/更新筛选和投资数据文件。 - 技能触发场景包括基金筛选、定投、分析、对比、组合建议等常见需求。
Metadata
Slug fund-screening
Version 1.0.0
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 1
Frequently Asked Questions

What is fund-screening?

基金筛选与定投实战技能。使用五维筛选体系(业绩/经理/风格/持仓/机构)从天天基金网和晨星网筛选优质基金,构建投资组合并执行定投策略。触发场景:用户提到"基金筛选"、"基金定投"、"选基"、"筛选基金"、"基金组合"、"定投策略"、"fund screening"、"fund DCA",或要求推荐/分析/对比基金... It is an AI Agent Skill for Claude Code / OpenClaw, with 92 downloads so far.

How do I install fund-screening?

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

Is fund-screening free?

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

Which platforms does fund-screening support?

fund-screening is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created fund-screening?

It is built and maintained by Allen Chen (@cr330326); the current version is v1.0.0.

💬 Comments