← Back to Skills Marketplace
jiayinian

jy-fund-manager-viewpoint

by jiayinian · GitHub ↗ · v1.0.2 · MIT-0
cross-platform ⚠ suspicious
76
Downloads
0
Stars
0
Active Installs
2
Versions
Install in OpenClaw
/install jy-fund-manager-viewpoint
Description
基于 MCP 聚源金融数据库分析基金定期报告中基金经理对特定行业的观点变化,对比不同季度的共识与分歧,输出带引用标注的结构化报告。使用场景:当用户需要分析某行业基金经理的观点演变、对比季度间共识与分歧、生成基金经理观点分析报告时触发。
README (SKILL.md)

基金经理观点分析 Skill

基于聚源 MCP 数据库获取基金定期报告,使用 OpenClaw 内置语义能力分析基金经理对特定行业的观点变化,生成带引用标注的结构化报告。

功能范围

  • ✅ 筛选目标行业主题基金(最多 15 只)
  • ✅ 获取基金定期报告全文(季度报告)
  • ✅ OpenClaw 语义提取基金经理对行业的观点
  • ✅ 识别季度内共识(至少 2 只基金的相似观点)
  • ✅ 识别季度内分歧(不同基金的不同看法)
  • ✅ 对比两个季度的观点变化(情绪、主题、策略)
  • ✅ 生成结构化 Markdown 报告(带引用溯源)

查询建议

查询要素: 行业名称、季度 1、季度 2、输出目录(可选)

查询写法:

分析 \x3C行业> 基金在 \x3C季度 1> 和 \x3C季度 2> 的观点变化
对比 \x3C季度 1> 和 \x3C季度 2> \x3C行业> 行业基金经理的观点
生成 \x3C行业> 行业 \x3C季度 1> vs \x3C季度 2> 基金经理观点分析报告

行业示例: 新能源、消费、人工智能、医药、金融、制造、周期、房地产

季度格式: 2025 年第三季度2025Q32024Q1

查询示例

# 分析新能源基金 2025 年 Q3 vs Q4 的观点变化
./run.sh "新能源" "2025Q3" "2025Q4"

# 分析消费基金 2024 年 Q1 vs Q2 的观点变化
./run.sh "消费" "2024Q1" "2024Q2"

# 指定输出目录
./run.sh "人工智能" "2025Q3" "2025Q4" "~/Desktop/reports"

环境检查与配置

每次使用本技能前,必须先检查 mcporter 安装和 MCP 服务配置状态!

步骤 1:检查 mcporter 是否安装

mcporter --version

如未安装,按以下流程安装:

# 1. 通过 npm 全局安装
npm install -g mcporter

# 2. 验证安装
mcporter --version

步骤 2:检查 MCP 服务配置

# 列出所有已配置的 MCP 服务
mcporter list

预期输出(必须包含以下两个服务):

  • jy-financedata-tool
  • jy-financedata-api

如服务未配置,需要获取 JY_API_KEY 并配置:

  1. 获取 JY_API_KEY,向恒生聚源申请 JY_API_KEY,通过邮箱申请(首次配置需提供,配置一次即可)

JY_API_KEY 申请路径: 向恒生聚源官方邮箱发送邮件申请签发 数据地图 JY_API_KEY,用于接口鉴权 申请通过后,恒生聚源将默认发送【工具版和接口版】KEY 另外,【Skill】包可通过 https://clawhub.ai/ 自行选择下载,若需要我们通过邮件提供【Skill】,亦可在邮件中注明

申请邮箱:[email protected]

邮件标题:数据地图 KEY 申请-XX 公司 - 申请人姓名

正文模板:

  • 姓名:
  • 手机号:
  • 公司/单位全称:
  • 所属部门:
  • 岗位:
  • MCP_KEY 申请用途:
  • Skill 申请列表:
  • 是否需要 Skill 安装包:(是,邮件提供/否,自行下载)
  • 其他补充说明(可选):
  1. 配置 MCP 服务
# 配置 jy-financedata-tool 服务
mcporter config add jy-financedata-tool --url "https://api.gildata.com/mcp-servers/aidata-assistant-srv-tool?token=你的 JY_API_KEY"

# 配置 jy-financedata-api 服务
mcporter config add jy-financedata-api --url "https://api.gildata.com/mcp-servers/aidata-assistant-srv-api?token=你的 JY_API_KEY"
  1. 验证配置
mcporter list
  1. 使用方式
# 基础键值对传参
mcporter call 服务名称。工具 参数=值

# 示例,注意:所有服务工具的入参均为 query
mcporter call jy-financedata-api.StockBelongIndustry query="电子行业 代表性上市公司 龙头股"

步骤 3:在 OpenClaw 中启用 mcporter(如未配置)

mcporter 配置文件路径

  • Windows: C:\Users\你的用户名\config\mcporter.json
  • Linux/MacOS: /root/config/mcporter.json

OpenClaw 配置文件路径

  • Windows: C:\Users\你的用户名\.openclaw\openclaw.json
  • Linux/MacOS: ~/.openclaw/openclaw.json

编辑 openclaw.json,在 skills 部分添加 mcporter 配置:

{
  "skills": {
    "entries": {
      "mcporter": {
        "enabled": true,
        "env": {
          "MCPORTER_CONFIG": "/root/config/mcporter.json"
        }
      }
    }
  }
}

重启 OpenClaw 使配置生效

openclaw gateway restart

核心工作流程

步骤 1:筛选目标行业基金

调用 jy-financedata-tool.SmartFundSelection 筛选行业主题基金(最多 15 只)。

mcporter call jy-financedata-tool.SmartFundSelection query="\x3C行业> 基金"

步骤 2:获取定期报告

调用 jy-financedata-api.FundAnnouncement 获取两个季度的季度报告全文。

mcporter call jy-financedata-api.FundAnnouncement query="\x3C基金名> \x3C年份>年第\x3C季度>季度报告"

步骤 3:OpenClaw 语义提取观点

将获取的报告内容提交给 OpenClaw,提取核心观点、关注主题、风险判断、情绪倾向。

步骤 4:OpenClaw 汇总季度观点

识别共识(至少 2 只基金相似观点)与分歧(不同基金不同看法)。

步骤 5:OpenClaw 对比季度变化

分析情绪变化、主题转移、新增担忧/机会、策略变化。

步骤 6:生成结构化报告

输出 Markdown 格式报告,包含核心结论、季度对比表格、基金观点详情、引用来源。

快速开始

基础用法

./run.sh "\x3C行业>" "\x3C季度 1>" "\x3C季度 2>"

工具调用命令

用途 命令
筛选行业基金 mcporter call jy-financedata-tool.SmartFundSelection query="\x3C行业> 基金"
获取基金报告 mcporter call jy-financedata-api.FundAnnouncement query="\x3C基金名> \x3C年份>年第\x3C季度>季度报告"
验证配置 mcporter list

数据获取策略

  1. 优先用基金名称查询,代码可能不识别
  2. 季度格式:2025 年第 3 季度2025Q3
  3. 必须使用 query 参数
  4. 季度报告在季度结束后 15 个工作日内披露

完整流程示例

# 1. 运行技能,准备数据
./run.sh "新能源" "2025Q3" "2025Q4"

# 2. 将生成的提示词发送给 OpenClaw 进行语义分析
# 提示词保存在 output/analysis_*.json 的 prompt 字段

# 3. 保存 OpenClaw 返回的 JSON 结果为 result.json

# 4. 生成最终报告
python3 scripts/generate_report.py output/analysis_*.json result.json

资源清单

~/Desktop/jy-fund-manager-viewpoint/
├── SKILL.md                  # 技能说明(本文件)
├── run.sh                    # 入口脚本
├── scripts/
│   ├── analyze_viewpoints.py # 核心分析脚本(数据准备 + 提示词生成)
│   └── generate_report.py    # 报告生成脚本
├── references/
│   ├── output_format.md      # 输出格式示例
│   └── mcp_guide.md          # MCP 调用详细指南
└── output/                   # 报告输出目录

限制

  • ✅ 数据源:恒生聚源基金公告
  • ✅ 语义分析:使用 OpenClaw 内置语义能力,无需额外 API
  • ✅ 引用标注:所有观点均标注来源基金及报告
  • ⚠️ 报告披露:季度结束后 15 个工作日内
  • ⚠️ 分析范围:默认最多 15 只基金
  • ⚠️ 内容长度:单篇报告最多 8000 字符
  • ⚠️ 分析耗时:约 5-10 分钟
  • ❌ 非投资建议:仅供参考

故障排查

Q: mcporter 未找到? A: 运行 npm install -g mcporter 安装。

Q: 未检测到聚源 MCP 服务配置? A: 按"步骤 2"配置 jy-financedata-tool 和 jy-financedata-api 服务。

Q: 没有 JY_API_KEY? A: 按"步骤 2"流程向恒生聚源申请 ([email protected])。

Q: 找不到某季度的报告? A: 季度报告可能尚未披露,请确认季度结束已超过 15 个工作日。

Q: 观点提取为空? A: 基金报告可能未提及该行业,尝试放宽行业关键词。

Q: 分析结果为空或格式错误? A: 确保 OpenClaw 返回的是有效 JSON 格式,如包裹在代码块中脚本会自动提取。

Usage Guidance
What to consider before installing/using this skill: - Credential handling: The skill needs a MCP API key (JY_API_KEY) configured into mcporter; the registry metadata does not declare this — expect to supply and store a token in mcporter.json. Only use an API key you trust and avoid pasting other secrets into mcporter or OpenClaw configs. - Configuration changes: The SKILL.md instructs editing OpenClaw config to enable mcporter and set MCPORTER_CONFIG. Editing agent config and enabling a tool increases the agent's access surface; only proceed if you understand and trust mcporter and the MCP service. - Install source: The skill recommends installing mcporter globally via npm. Verify the mcporter package (publisher, npm page, source code) before `npm install -g`, or consider installing in an isolated environment (container or dedicated VM) to reduce host exposure. - Code review: The included scripts use subprocess calls to `mcporter` and parse outputs, then save prompts and reports to ~/Desktop/...; there is no hidden network destination or obfuscated code in the provided files. Still, confirm there are no modifications to unrelated agent skills or system-wide files beyond the recommended config edits. - Least privilege: If possible, create a limited-scope API key for read-only access and place mcporter config in a restricted directory. Avoid using admin/root keys. Consider running the skill in an isolated environment if you are unsure about the mcporter package. - Metadata mismatch: Ask the skill maintainer (or registry) to update the declared requirements to include mcporter, node/npm, python3, and JY_API_KEY so the skill's metadata matches its behavior. If you need, I can: (1) point out the exact lines where credentials/config are referenced, (2) suggest safer installation steps (containerized), or (3) produce a short checklist to validate the mcporter npm package before installing.
Capability Tags
requires-sensitive-credentials
Capability Assessment
Purpose & Capability
The skill's stated purpose is to analyze fund manager views from the MCP (恒生聚源) database. The scripts and SKILL.md call mcporter and the jy-financedata tool/api services (jy-financedata-tool, jy-financedata-api) and instruct obtaining a JY_API_KEY — this is coherent with the declared purpose. However, the top-level registry metadata lists no required binaries or env vars while SKILL.md/openclaw metadata require node/npm/mcporter/python3 and a JY_API_KEY via mcporter; that mismatch is unexpected and should be clarified.
Instruction Scope
Runtime instructions and scripts operate within the stated scope: they call mcporter to list/configure services, fetch fund announcements, build prompts for OpenClaw, and generate Markdown reports. The SKILL.md also instructs editing OpenClaw configuration to enable mcporter and to place MCPORTER_CONFIG pointing at a mcporter JSON config — this is operationally required but broadens the agent's integration surface and should be acknowledged by the operator.
Install Mechanism
There is no packaged install spec in registry, but the SKILL.md and embedded openclaw metadata recommend installing mcporter via `npm install -g mcporter`. Global npm installs are common but execute third‑party code on the host; this is expected for the workflow, but operators should verify the mcporter package source and integrity before installing globally.
Credentials
The skill requires access to MCP credentials (JY_API_KEY) and the mcporter config file (mcporter.json) to call the MCP services, but the registry metadata does not declare any required environment variables or primary credential. The SKILL.md instructs adding MCPORTER_CONFIG to OpenClaw config and enabling the mcporter skill — that will expose a config file containing the API token to the agent environment. Requiring a single MCP API key is proportionate to the function, but the missing metadata declaration and instructions to edit agent config are notable security-relevant gaps.
Persistence & Privilege
The skill does not request always:true and does not itself persist credentials. It explicitly asks operators to enable the mcporter integration in OpenClaw (by editing openclaw.json) and to provide a mcporter config path; changing agent configuration is required for functionality but increases integration scope. Autonomous invocation is allowed (default) — combine this with credential access only if you trust the package and data source.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install jy-fund-manager-viewpoint
  3. After installation, invoke the skill by name or use /jy-fund-manager-viewpoint
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.2
- Added English description to the skill introduction for international usability. - Included a "资源清单" (resource structure) section listing all major files and directories. - Expanded and clarified troubleshooting steps for missing JY_API_KEY, missing reports, and analysis errors. - Stated report generation time (~5-10 minutes) in the limitations. - Minor formatting improvements and clearer section divisions in the documentation.
v1.0.1
- 新增详细使用说明与配置指引,包括聚源 MCP 服务与 mcporter 安装配置流程。 - 明确报告分析流程,实现对比两个季度基金经理对特定行业的观点变动,自动识别共识与分歧。 - 支持最多 15 只行业主题基金的筛选与报告批量处理。 - 完善命令行调用示例及数据处理流程,包括结构化 Markdown 报告输出,带明确信息溯源。 - 强化故障排查说明,便于环境配置与常见问题自助解决。
Metadata
Slug jy-fund-manager-viewpoint
Version 1.0.2
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 2
Frequently Asked Questions

What is jy-fund-manager-viewpoint?

基于 MCP 聚源金融数据库分析基金定期报告中基金经理对特定行业的观点变化,对比不同季度的共识与分歧,输出带引用标注的结构化报告。使用场景:当用户需要分析某行业基金经理的观点演变、对比季度间共识与分歧、生成基金经理观点分析报告时触发。 It is an AI Agent Skill for Claude Code / OpenClaw, with 76 downloads so far.

How do I install jy-fund-manager-viewpoint?

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

Is jy-fund-manager-viewpoint free?

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

Which platforms does jy-fund-manager-viewpoint support?

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

Who created jy-fund-manager-viewpoint?

It is built and maintained by jiayinian (@jiayinian); the current version is v1.0.2.

💬 Comments