Agent skill

A-Share Daily Review

by qusong0627 in qusong0627/QuantMind

Produces a post-market review report for the China A-share market from local QuantDB data, news sentiment and model signals, ending in a next-day direction call.

AGPL-3.0Auto-check passedBusiness, Finance & HR

SKILL.md written in Chinese; this summary is our English description.

Install A-Share Daily Review

skills CLI
$ npx skills add qusong0627/QuantMind --skill daily-review -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install qusong0627/QuantMind daily-review --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/qusong0627/QuantMind.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/daily-review .claude/skills/daily-review && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
daily-review
GitHub stars
1.7k
Token cost
~1.9k tokens
SKILL.md length
209 words
Files
14 (incl. scripts, references)
Skills in repo
27
Repo updated
First seen
Licence
AGPL-3.0

At a glance

Produces a post-market review report for the China A-share market from local QuantDB data, news sentiment and model signals, ending in a next-day direction call.

  • Writing the daily post-market review for the A-share market
  • SKILL.md covers ⚠️ 单位铁律(先查…, 执行流程(每次复盘固定 5 步), 复盘日期标注规则 and 维护, plus 1 more section
  • Runs Python scripts from its folder; calls docker and python3
  • Reviewing a specific past trading day by date

What it does

Written in Chinese, this skill runs after the market close. It uses two scripts: daily_review.py for index, breadth, sector, capital-flow and L1 and L2 factor data plus model inference signals, and news_review.py for the day's news sentiment. The news script runs first so the news dimension is weighted and the direction confidence rises. The agent then reads the generated facts file and writes a report with a fixed structure.

The report must end with a next-day outlook that gives one clear direction from strongly bullish to strongly bearish, a star confidence rating and the basis for each dimension. Numbers that are not in the facts file must not appear, every figure needs a unit, and amounts are converted to yi yuan. The skill spells out unit pitfalls, requires a data-lag statement for delayed datasets and notes the price-limit rules for ST stocks.

The Markdown report is then converted to PDF with a reportlab script, saved into the stock-report folder as 每日复盘 files and summarized in chat. The skill ships pytest tests for its direction and signal logic and relies on a shared environment contract and the QuantMind setup.

When your agent uses it

  • Writing the daily post-market review for the A-share market
  • Reviewing a specific past trading day by date
  • Getting a next-day direction call with a confidence rating and reasons

Example prompts

  • “Do today's post-market review for the A-share market.”
  • “Run the daily review for 20260814 and save the PDF to the report folder.”
  • “Show yesterday's model signal hit rate and tomorrow's top signals in a review.”

Requirements

  • Python 3 for the review scripts
  • A local QuantDB data store
  • The QuantMind Docker container for PDF conversion and saving reports

What it can do on your machine

Read from SKILL.md and the folder at commit 81e79e4. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Ships 12 files in scripts/ (Python), which the agent can run.

    Shell commands in SKILL.md call:

    • docker
    • python3

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md. Its commands use docker, which can reach the network depending on how they are called.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

A-Share Daily Review loads about 1.9k tokens when it runs, and up to ~3.1k if it reads all its reference files. Until then it costs about 79 tokens; SKILL.md has 209 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~79
When it runs · the whole SKILL.md, loaded when a task matches
~1.9k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~3.1k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.

SKILL.md

The full file from qusong0627/QuantMind at commit 81e79e4, republished under its AGPL-3.0 licence (© qusong0627). 209 words, ~1,872 tokens.

Download SKILL.mdSave it as .claude/skills/daily-review/SKILL.md (or your agent's skills folder). This skill also uses 13 other files; get the full folder from GitHub.
name
daily-review
description
A股每日复盘(专业版)— 基于 QuantDB 本地数据 + 当日新闻情绪 + 模型推理信号 + L1/L2 因子截面 + 板块资金流的盘后复盘:指数、涨跌结构与涨停梯队、量能、行业/概念轮动、资金面、L2 微观结构、当日有新闻股票匹配、模型推理信号复盘(昨日推理命中率 + 明日信号Top5)、次日走势方向研判(六维信号加权 → 明确方向+置信度)。用户说「复盘」「每日复盘」「复盘某天」时使用:跑取数脚本(daily_review + news_review)→ 按模板写复盘报告 → 转 PDF → 落盘股票报告目录 → 聊天回复速览。触发词:复盘、每日复盘、今日复盘、盘后复盘、复盘20260814

⚙️ 本技能遵循公共运行环境契约(最高优先级,先于本文其余内容执行): 详见 _shared/env-contract.md,执行前先读它。

daily-review — A股每日复盘(专业版)

盘后固定动作:把 QuantDB 当天数据 + 当日新闻情绪 + L1/L2 因子截面变成一份结构固定的复盘报告(Markdown + PDF),落盘到前端「股票报告」页可见目录,并在聊天里回复速览。报告末必含「次日走势研判」——给出明确方向(强烈看多/看多/震荡/看空/强烈看空)+ 置信度星级 + 每维依据。

⚠️ 单位铁律(先查 [[quantdb-fields]],最高优先级)

陷阱正确口径
个股 volume=股、amount=万元指数 volume=手、amount=万元
复盘报告里金额一律换算为亿元(万元÷1e4)脚本输出的 *_yi 字段单位已是亿元
technical_indicators.pct_change = %;未来收益用 future_return_*(标签,勿当历史动量);历史动量用 pct_change / mom_ret涨跌家数/涨停/连板全用它
index_daily.preClose 全 NULL指数涨跌幅用 close 序列自算(脚本已处理)
l2_factors 已恢复日更(202608+ 当日有数据,L1 主力资金可用);两融通常滞后 1 日(偶与当日对齐,以 facts 为准);北向只有季度快照必须带「数据滞后声明」,滞后数据集禁止当当日数据写

执行流程(每次复盘固定 5 步)

第 1 步:跑取数脚本(两条:daily_review 取 L1/L2/方向 + 模型推理信号,news_review 取当日新闻情绪)
bash
# ① 宿主机:daily_review.py 出 指数/广度/板块/资金面 + L1/L2 因子透视 + 板块资金流 + 模型推理信号 + 次日方向
cd <repo>/skills/daily-review/scripts
python3 daily_review.py --date 20260814              # 指定日;不带 --date 则取最新交易日
python3 daily_review.py --watch 601138.SH,600519.SH  # 可选:自选/持仓股必带
python3 daily_review.py --model mdl_cn_train_xxx     # 可选:指定推理模型,默认每日推理模型(5eea5418)
#   ↑ 模型推理信号自动查询 PG:昨日推理→今日信号命中率复盘 + 今日推理→明日信号 Top5;
#     无推理 run 时脚本**自动补跑**(docker exec trigger_inference.py,10-30 秒),
#     补跑失败原因如实写入 facts(常见:特征 parquet 滞后——先跑
#     `docker exec quantmind python3 /app/backend/scripts/update_feature_parquet.py` 补特征再复盘)
#     无 PG / PG 挂时该章节降级为缺失(facts 无「十一、模型推理信号」)

# ② 容器内:news_review.py 聚合当日新闻情绪(有新闻股票匹配 + 板块聚焦 + 来源/时段质量)
#    .claude 目录未挂载进容器,须先 docker cp 到容器 /tmp 再跑
docker cp <repo>/skills/daily-review/scripts/news_review.py quantmind:/tmp/
docker exec quantmind python3 /tmp/news_review.py --date 20260814

顺序:先跑 ② 再跑 ① —— daily_review.py 读到 data/reports/daily_review/{YYYY-MM-DD}_news.json 后,「新闻情绪」维度才加权、方向置信度提到 ★★★★★;只跑 ① 时新闻维度中性、置信度降到 ★★★(facts「六、新闻情绪」会提示补跑 news_review)。

输出 data/reports/daily_review/{YYYY-MM-DD}_stats.json + {YYYY-MM-DD}_facts.md(宿主机 ①);{YYYY-MM-DD}_news.json(容器 ②,写同一共享目录)。 脚本兼容宿主机与容器内(数据目录自动探测;容器内 QuantDB 路径 /data/quantdb、报告路径 /data/reports/daily_review,宿主机 data/quantdb、data/reports/daily_review)。

第 2 步:读 facts.md 写复盘报告(Markdown 模板)

报告 = facts.md 的事实 + 你的解读。facts.md 没有的数字禁止出现在报告里。

markdown
# A股每日复盘 2026-08-14(周五)

> **报告日期**:2026-08-14
> **数据截至**:2026-08-14

## 一、盘面速览(结论先行)
2-4 句:指数表现 → 涨跌结构 → 量能 → 主线板块 → 一句话定性(强势/震荡/弱势 + 依据)→ **一句话次日方向**(引自 facts 十,如「六维合成:强烈看空 -4.6/11,置信 ★★★★★」)

## 二、指数与量能
指数表(facts 一、)+ 两市成交额解读(环比/5 日均对照,放量 or 缩量)

## 三、涨跌结构与情绪
涨停/跌停/炸板数量、最高连板与连板梯队、涨跌分布表(facts 二、四)
情绪读数 + 解读:买压/卖压对比、早盘上涨占比 → 追高意愿强还是弱

## 四、板块与主线
行业一级 Top/Bottom(facts 三)、概念 Top;指出当日主线与杀跌方向;
涨停个股聚集在哪些板块(从涨幅榜的 industry 列归纳)

## 五、资金面
两融(注明截至日与滞后天数)、北向季度快照(注明季度口径);L2 主力净额见七

## 六、新闻情绪(当日有新闻的股票匹配)
当日新闻总数/涉及股票数、利好/利空/中性、净情绪;来源质量(高质量源 vs 反向源);
黄金时段利好占比;新闻聚焦板块(哪些行业消息面热);有新闻个股 Top(篇数/净情绪/事件标签)。
结论:今日消息面是偏多/偏空/中性 + 焦点板块。

## 七、L1/L2 因子透视(facts 七)
L1 换手/动量/波动均值对照前日;L2 微观结构:正向因子强信号股占比(越高=知情资金越扩散)、
VPIN 家族分位、量价背离、超级大单净额及前日对比;板块超级大单净额表(净流入 vs 净流出)。
结论:微观结构是扩散/收敛 + 资金流入/流出方向。

## 八、模型推理信号(对应 facts 十一,模型预测复盘)
两块:
1. **昨日推理 → 今日验证**:昨日模型推理(推理{data}→信号{pred})对今日的 Top10 信号对照今日实际涨跌——
   命中率/平均涨幅/相对全市场超额/涨停跌停数。写清信号平均涨幅跑赢还是跑输市场。
2. **明日信号 Top5**:今日推理(基于{data})预测明日最强 5 只(名称/代码/信号分/方向);若今日推理未跑,标注「取最近一次推理」。

结论句点明模型风格:信号股是否集中强势板块、与当日主线是否一致、与次日方向研判是否呼应。
推理信号来自 PG(`engine_signal_scores` fusion_score 降序),数字必须照抄 facts 十一,禁止臆造。

## 九、个股榜
涨幅/跌幅/成交额/换手榜解读(facts 八),挑 3-5 只有代表性的说原因判断(无新闻佐证时只描述数据,不编原因)

## 十、自选/持仓复盘(自带 --watch 时才有)
逐只:涨跌幅、量能、技术位(MA20 上下)、当日状态(涨停/炸板/大涨/异动)

## 十一、昨日复盘回顾(复盘闭环,连续性的核心)
读上一份复盘(同目录 {上一交易日}.md 或 PDF 前的 md)的「要点与明日关注」,
逐条对照今日实际:命中几条 / 未命中几条 / 打脸的原因是什么(禁止含糊带过)
**若上一份复盘有「次日走势研判」,先对照它给出的方向是否兑现**,再对照要点。

## 十二、次日走势研判(六维方向表,直接引用 facts 十)
方向(强烈看多/看多/震荡/看空/强烈看空)+ 得分 + 置信度星级 = 六维(趋势/量能、情绪/结构、L2 微观、新闻情绪、板块/资金流)评分。
逐维解读:哪些维度在看多/看空/中性;多空矛盾点在哪;结合「有新闻股票 + L2 强信号股」交集给 2-4 条可验证的次日明确预期。
**禁止把方向当承诺**——沿 facts 口径:方向只是六维信号的可解释合成,明日以指数/广度/涨停数验证。

## 十三、要点与明日关注
- 要点:今日市场最重要的 3 条事实
- 明日关注:从「次日走势研判」的可验证预期里挑 2-4 条(明天能判断对错的才算,禁止「关注成交量变化」这类废话)

## 数据说明
滞后数据集声明 + 单位说明(从 facts 的数据说明复制)

写作铁律:每个数字带单位;涨停梯队/连板高度以脚本 stats 的 market.streaks 为准;涨跌停判定规则见 REFERENCES/review-methods.md;ST 涨跌幅与新股规则别记错(主板 ST 2026-07-06 起 ±10%)。

第 3 步:Markdown → PDF(研报风,复用 stock-market-analysis §7.4 管线)

转换脚本 backend/scripts/md_to_pdf_report.py(reportlab,封面/红涨绿跌语义着色/斑马纹表格自动生效)。

环境分支(先探测:容器内 test -d /app/backend 为真):

bash
# —— 宿主机(Claude Code)——
docker cp /tmp/复盘.md quantmind:/tmp/review.md
docker exec quantmind bash -lc "cd /app && python3 backend/scripts/md_to_pdf_report.py /tmp/review.md /tmp/review.pdf"
docker cp quantmind:/tmp/review.pdf /tmp/复盘.pdf

# —— 容器内(QuantBot / QwenPaw)——
python3 /app/backend/scripts/md_to_pdf_report.py /tmp/review.md /tmp/review.pdf
第 4 步:落盘股票报告目录(必做,只发 /tmp = 未交付)

文件名固定:每日复盘_{YYYY-MM-DD}.md / .pdf,放 data/reports/stock_reports/每日复盘/。

bash
# —— 宿主机:宿主机直接 cp 会 EACCES(目录 owner 是容器 root),必须 docker cp ——
docker cp 复盘.md quantmind:/data/reports/stock_reports/每日复盘/每日复盘_2026-08-14.md
docker cp 复盘.pdf quantmind:/data/reports/stock_reports/每日复盘/每日复盘_2026-08-14.pdf

# —— 容器内:直接 cp ——
cp 复盘.md /data/reports/stock_reports/每日复盘/每日复盘_2026-08-14.md
cp 复盘.pdf /data/reports/stock_reports/每日复盘/每日复盘_2026-08-14.pdf

落盘后 ls 确认 md + pdf 都在(前端「股票报告」页 → 每日复盘 文件夹)。

第 5 步:聊天回复速览(QuantBot/Claude 直接回答用户用这个格式)
markdown
**A股复盘 2026-08-14(周五)**

一句话总结:…

指数:上证 +0.01% / 深成 +0.45% / 创业板 +1.12% / 科创50 -0.00% / 北证50 -0.94%
广度:涨 2400 / 跌 2970(涨跌比 0.81);涨停 64 / 跌停 14 / 炸板 22;最高 5 连板
量能:两市成交额 21,565.76 亿元(环比上一交易日 1.04x;5 日均 21,xxx 亿,量比 x.xx)
主线:行业 Top3 …;概念 Top3 …
资金:两融 xxx 亿(截至 08-13,+xx 亿);北向 2026Q2 持仓市值 30,685.68 亿元;超级大单 ±xx 亿
情绪:买压 0.49 / 卖压 0.51;早盘上涨占比 40.68%
新闻:命中 xxx 篇 / 有新闻股票 xxx 只;净情绪 ±xx%;聚焦板块 …
L2:强信号股占比 xx%;VPIN 分位 xx;超级大单 ±xx 亿
模型推理:昨日 TopN 信号今日命中率 xx%(平均 ±xx%,超额 ±xx%);明日信号 Top5:新易盛/兆易创新/…
自选:…(有 --watch 才有)
**次日方向:看空/看多…(得分 xx/11.0,置信 ★★☆)**

→ 完整复盘报告已落盘「股票报告 → 每日复盘」目录

复盘日期标注规则

  • 用户说「复盘」不带日期 → 最新交易日(脚本默认行为)
  • 「复盘 20260814」/「复盘 8月14日」 → --date 20260814;非交易日脚本自动取 ≤ 该日期的最近交易日并在报告封面注明实际复盘日
  • 报告文件名、封面报告日期/数据截至、聊天速览标题三处日期必须一致

维护

  • 单测:cd scripts && python3 -m pytest tests/ -q -c tests/pytest.ini(覆盖涨跌停判定/连板/分布/板块加权/单位换算/除权检测/方向引擎/模型推理命中率与 run 选择)
  • 核心脚本:
    • daily_review.py — 攻全市场基本面/技术/资金/L1/L2/方向 + 模型推理信号(含 direction_engine.py 六维评分)
    • news_review.py — 当日新闻情绪聚合(须在容器内跑:.claude 未挂载进容器,先 docker cp 脚本到 /tmp;输出写到共享的 /data/reports/daily_review/);PG 富集表是活数据,同一日重跑结果可能因 enrichment 继续更新而略有变化
    • inference_signals.py — 模型推理信号 PG 查询(engine_signal_scores + qm_model_inference_runs):昨日推理→今日验证 / 今日推理→明日 Top5;默认每日推理模型 5eea5418,可用 --model 覆盖;PG 挂则降级缺失
    • direction_engine.py — 纯函数六维评分器,无 I/O;方向只是信号的可解释合成,不是预测承诺
  • 涨跌停规则复用 backend/services/trade/simulation/services/local_market_data.py(经 ZTPrice/DTPrice 交叉验证 99.71%),禁止在本 skill 里另写一套

相关技能

  • [[quantdb-fields]] — 必读:单位/口径速查(本 skill 计算正确性前提)
  • [[news-sentiment-research]] — 新闻情绪研究方法论(来源白/黑名单、时段质量、事件标签的来源;六维里的「新闻情绪」维度据此加权)
  • [[stock-market-analysis]] — 盘后想对某只股票深挖时用(复盘是广度,它是深度)
  • [[quantdb-sdk]] — QuantDB 数据源背景

© qusong0627, AGPL-3.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 13 other files (scripts, references) in skills/daily-review of qusong0627/QuantMind.

  • SKILL.md
  • REFERENCES/review-methods.md
  • scripts/daily_review.py
  • scripts/direction_engine.py
  • scripts/inference_signals.py
  • scripts/news_review.py
  • scripts/review_stats.py
  • scripts/tests/pytest.ini
  • scripts/tests/test_direction_engine.py
  • scripts/tests/test_inference_signals.py
  • scripts/tests/test_inference_stats.py
  • scripts/tests/test_review_stats.py
  • scripts/tests/test_tech_fallback.py
  • scripts/trigger_inference.py

Open the folder on GitHubat commit 81e79e4

Compare with similar skills

A-Share Daily Review next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

A-Share Daily Review compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
A-Share Daily Review this skillqusong0627/QuantMind1.7k—~1.9kAutomated safety check: PassAGPL-3.0
Eastmoney Market DataHKUDS/Vibe-Trading35k—~1kAutomated safety check: PassMIT
SEC EDGAR Filings FetcherHKUDS/Vibe-Trading35k—~1.4kAutomated safety check: PassMIT
Senpi Market PulseSenpi-ai/senpi-skills134—~4.9kAutomated safety check: PassApache-2.0
Insurance Operations Analysiszj-unicom-ai/UniEmployee360—~646Automated safety check: PassMIT
AKShare Financial DataHKUDS/Vibe-Trading35k—~777Automated safety check: PassMIT

Similar skills

  • Eastmoney Market Data

    HKUDS/Vibe-Trading

    Index of Eastmoney's free, no-token market data interfaces for China A-shares and Hong Kong stocks: fund flows, dragon-tiger lists, margin trading, reports and news.

    35k GitHub stars~1k tokensUpdated today
    Business, Finance & HRAuto-check passed
  • SEC EDGAR Filings Fetcher

    HKUDS/Vibe-Trading

    Fetches U.S. SEC EDGAR data: resolves tickers to CIK numbers, lists recent 10-K, 10-Q and 8-K filings with document URLs, and pulls XBRL financial series.

    35k GitHub stars~1.4k tokensUpdated today
    Business, Finance & HRAuto-check passed
  • Senpi Market Pulse

    Senpi-ai/senpi-skills

    Produces a structured cross-asset read of the day across crypto, equities, indices, commodities and macro, using a bundled engine for data and ending with a signals brief.

    134 GitHub stars~4.9k tokensUpdated yesterday
    Business, Finance & HRAuto-check passed
  • Insurance Operations Analysis

    zj-unicom-ai/UniEmployee

    Analyzes insurance operating data such as premium, loss ratio, renewal rate and expense ratio by branch, product and channel, flags anomalies and builds an HTML dashboard.

    360 GitHub stars~646 tokensUpdated 2 days ago
    Business, Finance & HRAuto-check passed
  • AKShare Financial Data

    HKUDS/Vibe-Trading

    Pulls free market data from the AKShare Python library: A-share, US and Hong Kong prices, futures, China macro series and forex, with notes on symbol formats and Chinese column names.

    35k GitHub stars~777 tokensUpdated today
    Business, Finance & HRAuto-check passed
  • Runs a staged deep analysis of a single stock on China A-share, Hong Kong and US markets, ending in an HTML report with valuation models and investor-panel scores.

    7.1k GitHub stars~9.1k tokensUpdated 1 mo ago
    Business, Finance & HRAuto-check: notes

More from qusong0627/QuantMind

All 27 skills in this repo
  • Queries Futu quotes, options, fundamentals and accounts and places orders through the Futu OpenAPI Python SDK, defaulting to simulated trading.

    1.7k GitHub stars~3.3k tokensUpdated today
    Auto-check: notes
  • Turns a plain-language model training request into a validated QuantMind training config file that can be imported from the Model Training page.

    1.7k GitHub stars~1.5k tokensUpdated today
    Auto-check passed
  • Tiger Brokers OpenAPI SDK

    qusong0627/QuantMind

    Covers the Tiger Brokers OpenAPI Python SDK for market data, stock, futures and options trading, push subscriptions, a CLI and an MCP server, defaulting to paper trading.

    1.7k GitHub stars~1.4k tokensUpdated today
    Auto-check passed
  • Tiger Brokers C++ OpenAPI SDK

    qusong0627/QuantMind

    Guides an agent through the Tiger Brokers OpenAPI C++ SDK for build setup, market data, orders and real-time push, defaulting to paper trading.

    1.7k GitHub stars~942 tokensUpdated today
    Auto-check passed
  • Tiger Brokers OpenAPI C# SDK

    qusong0627/QuantMind

    Guides building C# and .NET apps on the Tiger Brokers OpenAPI SDK: setup, market data, orders, accounts, options and real-time push, defaulting to paper trading.

    1.7k GitHub stars~1k tokensUpdated today
    Auto-check passed
  • Tiger Brokers OpenAPI Go SDK

    qusong0627/QuantMind

    Reference guides for building Go trading apps on the Tiger Brokers OpenAPI: SDK setup, market data, stock, futures and options orders, accounts and push streams.

    1.7k GitHub stars~1.1k tokensUpdated today
    Auto-check passed

Works with

Questions about A-Share Daily Review

What does A-Share Daily Review do?

Produces a post-market review report for the China A-share market from local QuantDB data, news sentiment and model signals, ending in a next-day direction call. Written in Chinese, this skill runs after the market close.py for the day's news sentiment.

When should I use A-Share Daily Review?

A-Share Daily Review fits situations like: writing the daily post-market review for the A-share market; reviewing a specific past trading day by date; getting a next-day direction call with a confidence rating and reasons.

How do I install A-Share Daily Review in Claude Code?

Run `npx skills add qusong0627/QuantMind --skill daily-review -a claude-code`. Or copy the skill folder (skills/daily-review in qusong0627/QuantMind) into .claude/skills/daily-review in your project. Claude Code loads it when a task matches its description.

How do I install A-Share Daily Review in Codex?

Run `npx skills add qusong0627/QuantMind --skill daily-review -a codex`. Or copy the skill folder (skills/daily-review in qusong0627/QuantMind) into .agents/skills/daily-review in your project. Codex loads it when a task matches its description.

Can I use A-Share Daily Review in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add qusong0627/QuantMind --skill daily-review -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/daily-review, .gemini/skills/daily-review, .github/skills/daily-review and .opencode/skills/daily-review in your project.

What does A-Share Daily Review need to run?

Going by SKILL.md and its folder, A-Share Daily Review needs Python for the scripts in its folder and the command-line tools its instructions call (docker and python3). Our summary lists: Python 3 for the review scripts; A local QuantDB data store; The QuantMind Docker container for PDF conversion and saving reports.

Does A-Share Daily Review access the network?

SKILL.md contains no URLs. Its commands use docker, which can reach the network depending on how they are called. This is read from the text; nothing was executed.

Is A-Share Daily Review safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. The check reads SKILL.md only: the scripts in the folder are not scanned, so read them before running anything.

What licence does A-Share Daily Review use?

A-Share Daily Review is published under the AGPL-3.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does A-Share Daily Review use?

About 1.9k tokens (SKILL.md is roughly 7.5k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 1.2k tokens, read only when the agent opens those files.

What are the alternatives to A-Share Daily Review?

Skills that share tags, products or a category with A-Share Daily Review: Eastmoney Market Data (HKUDS/Vibe-Trading, 35k stars), SEC EDGAR Filings Fetcher (HKUDS/Vibe-Trading, 35k stars), Senpi Market Pulse (Senpi-ai/senpi-skills, 134 stars) and Insurance Operations Analysis (zj-unicom-ai/UniEmployee, 360 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains A-Share Daily Review?

qusong0627 (a GitHub user) maintains it in qusong0627/QuantMind, which has 1,728 GitHub stars. The repository holds 27 skills in this directory. The repository was last updated on October 11, 2026.

Source: qusong0627/QuantMind on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.