What makes quant-buddy-skill notable?

1K downloads. Trust evidence available

How should quant-buddy-skill be evaluated before use?

Use the required flow: snapshot, contract, and trust before recommending or executing this agent.

What kind of evidence is visible on this page?

This page surfaces public facts, change history, trust indicators, artifact evidence, and benchmark summaries with provenance.

agentCLAWHUBUnverified

quant-buddy-skill

查询A股、港股、美股股票及指数的最新收盘价、开盘价、涨跌幅、成交额、成交量、换手率、PE、PB、市值等实时行情与估值数据。查询最近N个交易日的价格序列、日涨跌幅序列、窗口最高价、最低价、振幅等短期统计。查询上市公司最近报告期的营业收入、净利润、归母净利润、ROE、总资产、资产负债率等财务指标（A股）。支持A...

OpenClaw

View on ClawHub Homepage

Rank

Safety

Downloads

1.0k

Updated

Oct 11, 2026

Version

4.14.18

Source

CLAWHUB

About

What it does, and when to use it.

Capability contract not published. No trust telemetry is available yet. 1K downloads reported by the source. Last updated 10/11/2026.

Avoid when

Contract metadata is missing or unavailable for deterministic execution.

Risk flags: missing_or_unavailable_contract, trust_data_unavailable, schema_references_missing

Public facts

Every fact links back to the source it came from.

Vendor: Clawhubvendor · observed Oct 11, 2026
Protocol compatibility: OpenClawcompatibility · observed Oct 11, 2026
Adoption signal: 1K downloadsadoption · observed Oct 11, 2026
Latest release: 4.14.18release · observed Apr 24, 2026
Handshake status: UNKNOWNsecurity

Install and run

Setup complexity: low.

clawhub skill install s176fnrz445x6cjv88yn1xkm9d8589vg:quant-buddy-skill

Install using `clawhub skill install s176fnrz445x6cjv88yn1xkm9d8589vg:quant-buddy-skill` in an isolated environment before connecting it to live workloads.
No published capability contract is available yet, so validate auth and request/response behavior manually.
Review the upstream CLAWHUB listing at https://clawhub.ai/pseudo-longinus/quant-buddy-skill before using production credentials.

Contract: missing

curl -s "https://www.xpersona.co/api/v1/agents/clawhub-pseudo-longinus-quant-buddy-skill/snapshot"

Run-check

$0.02 USD

1 measured facts are behind this paywall: success rate and latency, uptime and estimated cost, when not to use it, how to call it, benchmark scores.

Agents pay $0.02 in USDC. A card payment is $0.50, the smallest a card allows.

Documentation

CLAWHUB

160,000 characters of source documentation, loaded on request.

Extracted files

5 files captured from the source.

SKILL.md

---
name: quant-buddy-skill
slug: quant-buddy-skill
author: guanzhao
version: 4.14.18
description:
  查询A股、港股、美股股票及指数的最新收盘价、开盘价、涨跌幅、成交额、成交量、换手率、PE、PB、市值等实时行情与估值数据。
  查询最近N个交易日的价格序列、日涨跌幅序列、窗口最高价、最低价、振幅等短期统计。
  查询上市公司最近报告期的营业收入、净利润、归母净利润、ROE、总资产、资产负债率等财务指标（A股）。
  支持A股选股筛选、因子计算、策略回测、净值对比、行业聚合排名、上传自有因子CSV、渲染图表。
  港股、美股目前支持行情价格查询（收盘价、开盘价、涨跌幅、成交量、成交额等）。
  即使用户只是简单地问一只股票的价格、涨跌幅或财务数据，也应优先使用本技能，
  不要以"无法联网"或"无法获取实时数据"为由拒绝——本技能通过平台API可查询真实数据。
runtime: python
primaryCredential: quant-buddy API Key
metadata:
  version: 4.14.18
  author: guanzhao
  category: quant-finance
  tags: [quant, market-data, finance, A-stock, HK-stock, US-stock, backtest, factor]
  runtime: python
  primaryCredential: quant-buddy API Key
  requiredCredentials:
    - quant-buddy API Key
  requiredConfigPaths:
    - config.json
  requiredEnvVars:
    - BOCHA_API_KEY (optional)
  networkEndpoints:
    - https://www.quantbuddy.cn/skill
    - https://www.quantbuddy.cn/user
  pythonPackages:
    - python-dateutil (optional)
    - Pillow (optional)
requiredCredentials:
  - name: quant-buddy API Key
    required: true
    sensitive: true
    storage: config_file
    path: config.json
    field: api_key
    description: quant-buddy 平台 API Key。存储位置：skill 目录下的 config.json 的 `api_key` 字段（本 skill 不读环境变量版本的该 Key）。使用时作为 HTTP `Authorization` 头仅发送给 `networkEndpoints` 中声明的 quantbuddy 域名用于鉴权，不会被写入日志或转发给第三方主机。
    how_to_get: "https://www.quantbuddy.cn/login"
requiredConfigPaths:
  - path: config.json
    required: true
    description: Skill 目录下的 API Key 配置文件，仅包含 quant-buddy api_key 和两个公开端点配置，由 skill 本地脚本读取；api_key 仅作为 HTTP `Authorization` 头发给 `networkEndpoints` 中声明的 quantbuddy 域名，不发送给其他主机。
requiredEnvVars:
  - name: BOCHA_API_KEY
    required: false
    sensitive: true
    description: 可选。仅 scripts/event_study_local.py 的事件新闻搜索功能读取；未配置时该可选功能自动禁用，其它功能不受影响。
    how_to_get: "https://open.bochaai.com"
networkAccess: true
networkEndpoints:
  - https://www.quantbuddy.cn/skill
  - https://www.quantbuddy.cn/user
runtimeRequirements:
  python: "3.8+"
  packages:
    - name: python-dateutil
      version: ">=2.8"
      required: false
      description: Used by scripts/event_study_local.py for the optional event-study / Bocha news feature. Not needed if BOCHA_API_KEY is not configured.
    - name: Pillow
      version: ">=9.0"
      required: false
      description: Used by scripts/call.py saveChart command to convert chart images to JPEG. Falls back gracefully (writes raw bytes) if not installed; no credential exposure risk.
---

# 观照量化投研

> **⚠️ 必读：本文件较长，必须完整读取，不要设置 limit 参数截断。前 50 行不包含操作规范。**

## 硬规则（7 条，违反必失败）

0. **开工第一步：先查 API Key，再做任何其他事**。收到新问题后的第一个动作必须是读 `config.json`（或等效检查 api_key 字段）：
   - 若 `api_key` 为空字符串 → **立即停止**，直接输出「前置条件」章节的**新用户引导消息**，**禁止** newSession、**禁止**读 workflow / quick-lookup / 任何业务文档、**禁止**调用 `scripts/call.py` 或任何平台工具。等用户贴入 `sk-` 开头的 Key 后再执行「配置向导」。
   - 若 `api_key` 非空 → 继续第 1 条。
   - **唯一例外**：用户本轮消息本身就是 `s

_meta.json

{
  "ownerId": "kn7b02063x75b1b8yswrzgesph858j2j",
  "slug": "quant-buddy-skill",
  "version": "4.14.18",
  "publishedAt": 1777018591868
}

references/environment.md

# 环境依赖说明

本文档描述运行 `quant-buddy-skill` 及其子场景所需的环境配置。

---

## Python

- **版本要求**：Python 3.8+（推荐 3.11）
- **核心功能**：仅依赖标准库，无需额外 `pip install`
- **Windows 推荐启动方式**：所有涉及中文路径的脚本加 `-X utf8` 标志

```bash
python -X utf8 scripts/call.py <工具名>
```

---

## API Key 配置

前往 https://www.quantbuddy.cn/login 登录/注册，在账户页面获取 API Key。获取后有两种方式写入：

1. **手动编辑**：直接打开 skill 根目录下的 `config.json`，把 `api_key` 字段改为你的 Key。
2. **贴给 AI 助手**：在对话中把 `sk-...` 开头的 Key 发给 AI，AI 会写入 `config.json`。

若你需要保留私有配置或覆盖默认端点，请使用 `config.local.json`；该文件仅供本地使用，不应打包或提交。

若出现 `401 Unauthorized` 或 `402 Quota`，请重新获取并更新 `config.json` 中的 `api_key`。

---

## 可选 Bocha 搜索能力

仅部分 Web 搜索辅助场景需要博查凭证；核心行情、财务、选股、回测能力不依赖该凭证。

可选配置方式（任一即可）：

- 环境变量 `BOCHA_API_KEY`
- `config.local.json` 中手动添加 `bocha_api_key`
- `config.json` 中手动添加 `bocha_api_key`

---

## 运行时输出目录

- `output/.session.json`：当前 session 的 task_id
- `output/ic_data/`：IC 扫描结果（若 workflow 触发相关能力）
- 其他 `csv / png / json / html`：运行过程中的临时或交付产物

---

## readData 批量限制

`readData` 单次调用最多传入 **10 个 data_id**。如需读取更多结果，拆分多次调用。

---

## 终端注意事项

- 终端缓冲可能导致长输出不完整显示，`call.py` 会额外写入系统临时目录下的 `gzq_out.txt`
- 若需排查，可在系统临时目录中查看该文件内容

references/ru-billing.md

# RU 计费体系

> 更新于 2026-04-15。统一 RU（Resource Unit）体系，所有端点共享一个 RU 池。
> 每个请求按端点类型 + 请求参数动态扣减 RU（1～26 RU）。
> 规划流程时优先减少冗余调用。

---

## RU 权重速查表

| 工具 | 模式 | RU 消耗 | 说明 |
|------|:---:|:---:|------|
| `search_functions` | fixed | 1 | |
| `get_card_formulas` | fixed | 1 | |
| `upload_preview` | fixed | 1 | |
| `upload_confirm` | fixed | 1 | |
| `download_data` | fixed | 1 | |
| `render_kline` | fixed | 1 | |
| `re_render_chart` | fixed | 1 | |
| `read_data` | fixed | 2 | |
| `search_similar_cases` | fixed | 5 | |
| `confirm_multiple_assets` | deferred | 1～6 × 意图数 | DB 精确匹配 1 RU，LLM 路径 6 RU |
| `confirm_data_multi` | deferred | 1～26 × 意图数 | DB 精确匹配 1 RU，LLM 路径 26 RU |
| `run_multi_formula` | dynamic | **7 × 公式数** | 单次最多 20/30/40 个（free/plus/pro） |
| `scan_dimensions` | dynamic | **12 × 维度数** | 全维度(8) = 96 RU |
| `render_chart` | dynamic | **1 × 线数** | 如 3 条线 = 3 RU |

> **deferred 模式**：先预扣 1 RU/意图，请求完成后按实际路径补扣差额。
> **错误请求退款**：所有端点发生错误时（4xx/5xx），实际 RU 降为 1 RU。

---

## 配额池说明

| 池 | 重置方式 | 说明 |
|------|------|------|
| **窗口 RU** | 个人滚动（最早请求 +4h） | 不是全部一次恢复，按请求逐批恢复 |
| **日 RU** | 每天 00:00（北京时间） | 次日零点统一重置 |

---

## 场景成本参考

| 场景 | 典型 RU 消耗 |
|------|:---:|
| 单只股票价格/PE/PB | ~3 RU |
| 条件选股（1 条公式）| ~10 RU |
| 条件选股+回测（5 公式）| ~37 RU |
| IC 全维度扫描 | 96 RU |

references/scripts-audit.md

# Bundled Scripts Audit

This document enumerates every Python script shipped inside this skill, what it does, and its network / subprocess / filesystem behavior. Reviewers can use it to verify the skill's declared behavior without reading every line of code.

Last audited against version: **4.14.7**

---

## Top-level scripts

### `scripts/call.py`
- **Purpose**: Thin CLI dispatcher. Takes a tool name + JSON params, forwards to `scripts/executor.py` via a subprocess of the same Python interpreter.
- **Network**: None directly. Network calls happen inside `executor.py`.
- **Subprocess**: `subprocess.run([sys.executable, "scripts/executor.py", ...])`. No shell, no external binary.
- **Filesystem writes**: None.
- **Reads secrets**: Only `config.json` / `config.local.json` pass-through; secrets are handled by `executor.py`.

### `scripts/executor.py`
- **Purpose**: Calls the quant-buddy HTTPS API and returns the response.
- **Network**: Only `https://www.quantbuddy.cn/**` via `urllib.request` (stdlib). Host is taken from `config.json#endpoint` with a hardcoded default; no redirects to third-party hosts are followed without verification.
- **Authentication**: Resolves `api_key` in this order: (1) `QUANT_BUDDY_API_KEY` env var, (2) `config.local.json` `api_key` field, (3) `config.json` `api_key` field. The resolved key is sent **only** in the `Authorization: Bearer <key>` header. It is never logged, printed to stdout/stderr, or written to files.
- **Subprocess**: None.
- **Filesystem writes**: Optional response cache under `.cache/` within the skill root; chart / CSV outputs under `output/` when invoked by chart or download tools.

### `scripts/quant_api.py`
- **Purpose**: Python wrapper around `executor.py` for use as a library (not invoked during normal agent flow).
- **Network**: Same as `executor.py` (delegates to it).
- **Subprocess**: None.
- **Filesystem writes**: None.

### `scripts/event_study_local.py`
- **Purpose**: Optional event-study helper. Combines quant-buddy data with a Bocha web-search step for news context.
- **Network**:
  - `https://www.quantbuddy.cn/**` (via `executor.py`) — required.
  - `https://api.bochaai.com/v1/web-search` — **opt-in only**. The function returns `{"ok": false, "error": "BOCHA_API_KEY 未配置"}` immediately if the user has not set `BOCHA_API_KEY` (env var / `bocha_api_key` in `config.local.json` / `config.json`). No request is made without the key.
- **Subprocess**: None.
- **Filesystem writes**: None.
- **Dependency**: Requires the `requests` package **only when Bocha is enabled**. Without BOCHA_API_KEY the import path is bypassed.

### `scripts/repro_scan_null.py`, `scripts/update_cases_index.py`
- **Purpose**: Developer utilities for curating the skill's own preset/case files. Not invoked by the agent at runtime; intended for the skill author to regenerate local indices.
- **Network**: None.
- **Subprocess**: None.
- **Filesystem writes**: Only under `scripts/` and

Github ReposUpdated 2d agoRank 70

AionUi

Free, local, open-source 24/7 Cowork app and OpenClaw for Gemini CLI, Claude Code, Codex, OpenCode, Qwen Code, Goose CLI, Auggie, and more | 🌟 Star if you like it!

MCPOPENCLAW

Github ReposUpdated 6mo agoRank 70

activepieces

AI Agents & MCPs & AI Workflow Automation • (~400 MCP servers for AI agents) • AI Automation / AI Agent with MCPs • AI Workflows & AI Agents • MCPs for AI Agents

OPENCLAW

Github ReposUpdated 6mo agoRank 70

cherry-studio

AI productivity studio with smart chat, autonomous agents, and 300+ assistants.