ClawHub

Skills of A-share fund data released by ft.tech.

v1.0.0

基金数据技能集（market.ft.tech）。覆盖基金基本信息、净值历史、累计收益率、基金概览列表、支持基金标的列表。用户询问基金详情、基金净值、基金收益、基金列表时使用。

⭐ 1· 273·0 current·0 all-time
by@shawn92

Install

OpenClaw Prompt Flow

Install with OpenClaw

Best for remote or guided setup. Copy the exact prompt, then paste it into OpenClaw for shawn92/ftshare-fund-data.

Previewing Install & Setup.
Prompt PreviewInstall & Setup
Install the skill "Skills of A-share fund data released by ft.tech." (shawn92/ftshare-fund-data) from ClawHub.
Skill page: https://clawhub.ai/shawn92/ftshare-fund-data
Keep the work scoped to this skill only.
After install, inspect the skill metadata and help me finish setup.
Use only the metadata you can verify from ClawHub; do not invent missing requirements.
Ask before making any broader environment changes.

Command Line

CLI Commands

Use the direct CLI path if you want to install manually and keep every step visible.

OpenClaw CLI

Bare skill slug

openclaw skills install ftshare-fund-data

ClawHub CLI

Package manager switcher

npx clawhub@latest install ftshare-fund-data
Security Scan
VirusTotalVirusTotal
Pending
View report →
OpenClawOpenClaw
Benign
high confidence
✓
Purpose & Capability
Name/description (A-share fund data from market.ft.tech) matches the included sub-skills and required actions: each handler issues HTTPS requests to market.ft.tech fund API endpoints and returns JSON. No unrelated credentials, binaries, or config paths are requested.
✓
Instruction Scope
SKILL.md and run.py restrict behavior to selecting a sub-skill and executing its handler; sub-skill SKILL.md files describe only parameterized queries. Handlers only build URLs and perform HTTPS GETs to market.ft.tech; they do not read local secrets, system files, or transmit data to other endpoints.
✓
Install Mechanism
No install spec — instruction-only with bundled Python scripts. Nothing is downloaded or written during install; code is executed locally via runpy.
✓
Credentials
No environment variables, credentials, or external service tokens are required. The handlers only use command-line args and make outbound HTTPS requests to the declared BASE_URL.
✓
Persistence & Privilege
always is false and the skill does not request persistent system-wide privileges or modify other skills. It runs as ordinary code when invoked.
Assessment
The skill appears coherent and limited to querying https://market.ft.tech for fund data. Before installing, confirm you trust the market.ft.tech domain (this skill will make outbound HTTPS requests to it) and that the source code bundle is from a trusted publisher (homepage is missing). If you operate in a restricted environment, consider running it in an isolated environment because it executes bundled Python scripts locally. Finally, avoid providing sensitive secrets as query parameters—this skill only needs fund codes and pagination parameters.

Like a lobster shell, security has layers — review code before you run it.

latestvk974qrmrqnzkcm4rzj0gk1z1wd83gr6w
273downloads
1stars
1versions
Updated 1mo ago
v1.0.0
MIT-0
@shawn92
latest
Download

FT 基金数据 Skills

本 skill 是 FTShare-fund-data 的统一路由入口。

根据用户问题，从下方「能力总览」或「询问方式与子 skill 对应表」匹配对应子 skill，然后通过 run.py 执行并解析响应。

所有接口均以 https://market.ft.tech 为基础域名。

调用方式（唯一规则）

run.py 与本文件（SKILL.md）位于同一目录。执行时：

  1. 取本文件的绝对路径，将末尾 /SKILL.md 替换为 /run.py，得到 <RUN_PY> 。
  2. 调用：python <RUN_PY> <子skill名> [参数...]
# 示例（<RUN_PY> 为实际绝对路径）
python <RUN_PY> fund-basicinfo-single-fund --institution-code 000001
python <RUN_PY> fund-cal-return-single-fund-specific-period --institution-code 159619 --cal-type 1Y
python <RUN_PY> fund-nav-single-fund-paginated --institution-code 000001 --page 1 --page-size 50
python <RUN_PY> fund-overview-all-funds-paginated --page 1 --page-size 20
python <RUN_PY> fund-support-symbols-all-funds-paginated --page 1 --page-size 20

run.py 内部通过 __file__ 自定位，无论安装在何处都能正确找到各子 skill 的脚本。

基金 — 询问方式与子 skill 对应表

询问方式（用户常说的词）子 skill
基金基本信息、某只基金详情、基金管理人/基金经理、基金类型/投资目标fund-basicinfo-single-fund
基金累计收益率、近1年/近3个月收益、YTD 收益、基金收益曲线fund-cal-return-single-fund-specific-period
基金净值、单位净值/累计净值、日增长率、基金净值历史fund-nav-single-fund-paginated
基金概览、所有基金信息、基金列表概况fund-overview-all-funds-paginated
支持的基金列表、基金代码清单、所有基金标的fund-support-symbols-all-funds-paginated

能力总览

  • fund-basicinfo-single-fund：查询指定基金基础信息（名称、管理人、经理、类型、投资目标等）。必填：--institution-code（6 位基金代码）。若用户仅给基金名称，先通过 fund-support-symbols-all-funds-paginated 或 fund-overview-all-funds-paginated 映射代码再查。
  • fund-cal-return-single-fund-specific-period：查询指定基金在指定区间的累计收益率时间序列。必填：--institution-code、--cal-type（1M/3M/6M/1Y/3Y/5Y/YTD）。建议先完成名称到代码映射后再调用。
  • fund-nav-single-fund-paginated：查询指定基金净值历史（分页）。必填：--institution-code；可选：--page、--page-size。建议先完成名称到代码映射后再调用。
  • fund-overview-all-funds-paginated：查询所有基金概览信息（分页）。可选：--page、--page-size。
  • fund-support-symbols-all-funds-paginated：查询所有支持基金的标的列表（分页）。可选：--page、--page-size。

使用流程

  1. 记录本文件绝对路径，将 /SKILL.md 替换为 /run.py 得到 <RUN_PY>。
  2. 理解用户意图，从「询问方式与子 skill 对应表」或「能力总览」匹配子 skill 名称。
  3. 若用户给的是基金名称而非代码：先调用 fund-support-symbols-all-funds-paginated 或 fund-overview-all-funds-paginated 获取候选，确定 6 位 institution-code。
  4. （可选）读取 sub-skills/<子skill名>/SKILL.md 了解接口与参数。
  5. 执行：python <RUN_PY> <子skill名> [参数...]，获取 JSON 输出（详情/净值/收益统一使用 --institution-code）。
  6. 解析并输出：以表格或要点形式展示给用户；若候选不唯一，先让用户确认再查询指标。

Comments

Loading comments...