据 AlphaMoat 统计,截至 2026-09,Tvscreener 累计下载 2,775 次、获 0 个星标,在全部 5,078 个 AI 技能中排名第 3,892 位,在「专业领域」分类(465 个)中排名第 326 位。

累计安装

208

累计下载

2,775

星标

全站排名

#3,892

全站前 77% · 「专业领域」前 71%

平台	累计安装	累计下载	星标	评论
ClawHub ↗	82	2,772	-	-
SkillHub ↗	208	2,775	-	—

各平台的计数口径不同,统计时间不一致。

这个技能是做什么的

适用于股票查询、技术指标（价格、涨跌幅、RSI、MACD、成交量）、标的过滤与自定义字段筛选。安装为 python3 -m pip install -U tvscreener，要求 Python 3.10 及以上。三个脚本入口：query_symbol.py 输出单标的预设技术快照（如 --symbol HKEX:700 --market HONGKONG）；run_query.sh 支持自定义字段与过滤器组合；discover_fields.py 按关键词发现可用字段。核心指标集涵盖 PRICE、CHANGE_PERCENT、VOLUME、RELATIVE_STRENGTH_INDEX_14、MACD(12/26) 三件套、SMA/EMA 20/50/200、布林带上下轨、随机指标 %K/%D、AVERAGE_TRUE_RANGE_14 与均线评级；过滤器支持 =、!=、>、<、>=、<=。已知缺陷：FIELD|60 等周期字段在部分 tvscreener 版本的 custom_query.py 会触发 FieldWithInterval 属性错误，规避方式是不带周期字段或改用 query_symbol.py。故障排查覆盖 ImportError（多 Python 环境不一致）与未加引号导致的 shell 管道误解析。回归测试脚本覆盖腾讯（港股）、茅台（A 股）、A 股 ETF 510300 与 BIDU（美股）。配套 skill-card 提醒：shell 包装可能在运行时静默安装或升级未锁版本的 Python 包，应在隔离虚拟环境运行、审查自动安装行并优先使用锁定的依赖清单。

资料发布于 2026-10-27

可以用来做什么

查询单只股票的技术指标快照
按字段与过滤器做市场筛选
按关键词发现可用字段

什么时候用

需要 TradingView 筛选器快照、技术指标或按条件过滤港股/A 股/ETF/美股时

示例提示词

以下是根据技能文档生成的示例,不代表实际运行结果。

查询 SHSE:600519 的技术指标快照

筛出 RSI 低于 30 的 A 股标的

月度数据趋势

累计下载2,775

星标-

2026-09 起统计方法调整:同名技能按作者分别统计,并覆盖全部技能。图中虚线为调整前的月份,与之后的数据不直接比较,也不计算跨越调整的增长。

月份	累计下载	环比	星标
2026-09	2,775	—	-
2026-08	2,637	+6.5%	-
2026-07	2,476	+13.3%	-
2026-06	2,186	+8.3%	-
2026-05	2,018	—	-
2026-03	1,419	—	-

SKILL.md 原文

以下是作者发布的原始文件,按纯文本显示,本站未做修改或审核;其中的链接不可点击,命令请核实后再运行。抓取于 2026-10-23。

3.2 KB · ClawHub v1.0.0

---
name: tvscreener
description: Query TradingView screener data for HK, A-share, A-share ETF, and US symbols with deepentropy/tvscreener. Use for stock lookup, technical indicators (price/change/RSI/MACD/volume), symbol filtering, and custom field/filter-based market queries.
---

# tvscreener

Use this skill for market queries with simple scripts first, then native Python when needed.

## Install

展开全部(共 79 行)


```bash
python3 -m pip install -U tvscreener
```

Python must be `>=3.10`.

## Quick commands (run from skill root)

> Use Python 3.10+ in your preferred environment (venv/pyenv/system Python).

```bash
# Preset single-symbol output (recommended)
python3 scripts/query_symbol.py --symbol HKEX:700 --market HONGKONG

# Custom query (fields + filters)
bash scripts/run_query.sh \
  --market CHINA \
  --symbol SHSE:600519 \
  --fields 'NAME,PRICE,CHANGE_PERCENT,VOLUME,RELATIVE_STRENGTH_INDEX_14,MACD_LEVEL_12_26,MACD_SIGNAL_12_26,MACD_HIST,SIMPLE_MOVING_AVERAGE_20,SIMPLE_MOVING_AVERAGE_50,SIMPLE_MOVING_AVERAGE_200,EXPONENTIAL_MOVING_AVERAGE_20,EXPONENTIAL_MOVING_AVERAGE_50,EXPONENTIAL_MOVING_AVERAGE_200,BOLLINGER_UPPER_BAND_20,BOLLINGER_LOWER_BAND_20,STOCHASTIC_PERCENTK_14_3_3,STOCHASTIC_PERCENTD_14_3_3,AVERAGE_TRUE_RANGE_14,MOVING_AVERAGES_RATING' \
  --filter 'NAME=600519'

# Field discovery
python3 scripts/discover_fields.py --keyword macd --limit 20
```

### Shell quoting notes

- Wrap `--fields` and `--filter` in single quotes.
- If you use interval syntax like `FIELD|60`, quoting is mandatory to avoid shell pipe parsing.


## Query rules

- Core technical set (recommended): `PRICE`, `CHANGE_PERCENT`, `VOLUME`, `RELATIVE_STRENGTH_INDEX_14`, `MACD_LEVEL_12_26`, `MACD_SIGNAL_12_26`, `MACD_HIST`, `SIMPLE_MOVING_AVERAGE_20/50/200`, `EXPONENTIAL_MOVING_AVERAGE_20/50/200`, `BOLLINGER_UPPER_BAND_20`, `BOLLINGER_LOWER_BAND_20`, `STOCHASTIC_PERCENTK_14_3_3`, `STOCHASTIC_PERCENTD_14_3_3`, `AVERAGE_TRUE_RANGE_14`, `MOVING_AVERAGES_RATING`
- Interval fields syntax: `FIELD|60` / `FIELD|240` (example: `RELATIVE_STRENGTH_INDEX_14|60`)
  - **Current caveat**: interval fields may fail in `scripts/custom_query.py` with `FieldWithInterval` attribute errors in some tvscreener versions.
  - Workaround: run without interval fields, or use `scripts/query_symbol.py` for stable single-symbol technical snapshots.
- Filters: `=`, `!=`, `>`, `<`, `>=`, `<=`

## Troubleshooting

- `ImportError: cannot import name 'Market' from 'tvscreener'`
  - Usually caused by mismatched Python/site-packages or multiple Python environments.
  - Fix: ensure commands and installation use the same Python (3.10+), then reinstall:
    - `python3 -m pip install -U tvscreener`
- `zsh: command not found: 60,...`
  - Cause: unquoted `FIELD|60` interpreted as shell pipes.
  - Fix: single-quote the full `--fields` string.

## References

- Workflow + patterns: `references/README_USAGE.md`
- API details:
  - `references/api/screeners.md`
  - `references/api/fields.md`
  - `references/api/filters.md`
  - `references/api/enums.md`

If scripts are insufficient, read references and write direct Python using tvscreener native API.

## Regression test

```bash
bash scripts/test_markets.sh
```

Covers Tencent (HK), Moutai (A), A-share ETF (510300), and BIDU (US).