install
source · Clone the upstream repo
git clone https://github.com/shaoxing-xie/openclaw-data-china-stock
Claude Code · Install into ~/.claude/skills/
T=$(mktemp -d) && git clone --depth=1 https://github.com/shaoxing-xie/openclaw-data-china-stock "$T" && mkdir -p ~/.claude/skills && cp -r "$T/skills/fund-flow-analyst" ~/.claude/skills/shaoxing-xie-openclaw-data-china-stock-fund-flow-analyst && rm -rf "$T"
OpenClaw · Install into ~/.openclaw/skills/
T=$(mktemp -d) && git clone --depth=1 https://github.com/shaoxing-xie/openclaw-data-china-stock "$T" && mkdir -p ~/.openclaw/skills && cp -r "$T/skills/fund-flow-analyst" ~/.openclaw/skills/shaoxing-xie-openclaw-data-china-stock-fund-flow-analyst && rm -rf "$T"
manifest:
skills/fund-flow-analyst/SKILL.mdsource content
Fund Flow Analyst
目标
对 A 股资金流进行多层结构化解读,识别主线资金抱团与分歧信号。
输入
- 用户问题
- 大盘/板块/个股资金流数据
- 北向资金数据
输出(固定结构)
- 大盘资金状态
- 北向资金态度
- 板块与个股资金聚集
- 连续性与反证
强制规则
- 对“单日”与“连续性(3/5/20日)”结论分开表达。
- 无法构成连续性证据时输出
。insufficient_evidence - 禁止输出买卖点、仓位比例、杠杆建议。
- 阈值从
读取。config/fund-flow-analyst_config.yaml
依赖工具
tool_fetch_a_share_fund_flowtool_fetch_northbound_flowtool_fetch_sector_datatool_fetch_market_data
通用输出字段
market_flownorthbound_flowsector_flow_rankcrowding_signalscontinuity_metricsimplicationscounterevidence
(low/medium/high)confidence_band