"i问财选股技能"

根据输入的链接、文件、文本或关键词分析利好行业/概念/题材/产业链，聚焦核心受益方向，在i问财网站进行选股查询，返回每项前5个结果并生成Excel文件。当用户要求分析利好行业、概念选股、行业选股、选股、利好哪些股票等意图时使用此技能。

Security Scan

The skill's stated purpose (query i问财 and generate Excel with top-5 stocks per concept) is plausible, but there are mismatches and small-but-meaningful issues (undeclared dependencies, a hard-coded user path, and a buggy fallback in the Excel script) that should be clarified before installing.

Purpose & Capability

The skill's name/description match the behavior: analyze input (link/file/text/keywords), query iwencai.com, capture top-5 results, and produce an Excel. However the README and script reference runtime dependencies (Python, pandas, openpyxl, and a web browser) that are not declared in the skill metadata or install spec — this is an implementation mismatch that should be justified.

Instruction Scope

SKILL.md stays within the stated scope (use web_fetch for links, read user-provided files, run searches on iwencai, aggregate results, generate Excel). Two concerns: (1) the SKILL.md directs saving files to a hard-coded path (/Users/tututu/.openclaw/workspace/...), which may be inappropriate or non-existent on other hosts and leaks an author-specific path; (2) the activation rule allows the model to '自行根据语义判断需要选股的场景' which gives broad discretion — acceptable but worth noting as it may trigger the skill in ambiguous contexts.

Install Mechanism

No install spec (instruction-only) — lower risk. But README lists dependencies (Python/pandas/openpyxl/Chrome or Firefox) that are not declared in metadata. The included script expects pandas and openpyxl (via pandas ExcelWriter engine), and the README's browser requirement suggests scraping/automation not described in SKILL.md. This mismatch should be resolved.

Credentials

The skill requests no environment variables, credentials, or config paths. There is no evidence it attempts to access unrelated secrets. The only file access described is user-supplied content and writing an output file to the workspace.

Persistence & Privilege

always:false and no special persistence requested. The skill does write an output file within a workspace (normal). It does not request to modify other skills or system-wide settings.

What to consider before installing

This skill appears to do what it claims (scrape/query i问财 and export top-5 results to Excel) but has implementation mismatches you should resolve before use: - Dependencies: README & script require Python, pandas, and openpyxl (and mention a browser) but the skill metadata declares no installs — confirm the runtime will provide these, or update the skill to declare/install them. - Bug: the script's ImportError fallback is broken (it prints a fallback message but then still references pandas, which will crash if pandas is absent). Ask the author to fix the fallback or ensure pandas is available. - Hard-coded path: it writes to /Users/tututu/.openclaw/workspace/... — verify this is acceptable for your environment or request a configurable output path. - Web access: the skill relies on web_fetch to contact iwencai.com; check whether iwencai requires special authentication, rate limits, or anti-scraping measures (the README's browser note hints at possible scraping complexity). - Autonomy: the skill can be invoked by the model when it infers a matching user intent — if you want stricter control, only enable it for explicit user requests. Recommend testing in a safe environment, requesting the developer to fix the script bug and to declare/install dependencies, and confirming the expected workspace path and network access before enabling for general use.

i问财选股技能 (iwencai-screener)

触发条件

当用户具有以下意图时激活：

• "XX选股" - 如"新能源选股"、"AI选股"
• "XX利好哪些股票" - 如"芯片利好哪些股票"
• "分析利好哪些行业"
• "概念选股"、"行业选股"
• "根据XX选股"
• 输入链接/文本后要求分析并选股
• 模型自行根据语义判断需要选股的场景

核心原则

聚焦原则：识别3-5个最核心的利好方向，避免泛泛而谈

• 优先选择与内容最直接相关的行业/概念
• 产业链分析聚焦核心环节，不做全产业链罗列
• 每个方向都要有明确的利好逻辑

操作流程

第一步：分析内容识别核心利好方向

1. 分析输入内容：

   • 链接：使用web_fetch获取内容并分析
   • 文件：读取文件内容并分析
   • 文本/关键词：直接分析
   • 语义判断：根据用户输入自行推断选股意图

2. 聚焦识别（只选最核心的3-5个方向）：

   选择标准：

   • 与内容直接相关（不是间接关联）
   • 有明确的产业链传导逻辑
   • 有实际的产品/业务支撑

   聚焦格式：

   【核心利好方向】（限定3-5个）
   
   1. XX行业/概念
      - 利好逻辑：为什么利好（具体原因）
      - 产业链位置：上游/中游/下游
      - 代表标的：YYY、ZZZ
   
   2. XX行业/概念
      - 利好逻辑：为什么利好
      - ...
   

3. 输出示例：

   【核心利好方向】
   
   1. 脑机接口（核心直接利好）
      - 利好逻辑：新闻提到脑机接口技术突破直接带动康复医疗器械需求
      - 产业链位置：下游应用端（康复设备）+ 中游制造
      - 代表标的：翔宇医疗、麦澜德
   
   2. 芯片半导体（上游支撑）
      - 利好逻辑：脑机接口需要专用芯片，刺激芯片需求
      - 产业链位置：上游核心零部件
      - 代表标的：国科微、紫光国微
   

第二步：i问财选股查询

4. 聚焦查询（只查核心方向）：

   • 访问：https://www.iwencai.com
   • 针对每个核心识别出的行业/概念进行搜索
   • 搜索示例：
     • "XX行业股票" 或 "XX概念股票"
     • "XX龙头股"

5. 获取查询结果：

   • 每条查询取前5个结果
   • 重要：获取所有返回字段，完整保留数据

第三步：汇总结果

6. 整合数据：

   • 按核心方向分类展示
   • 每个方向显示前5只股票

7. 生成Excel文件：

   • 使用generate_stock_excel.py生成
   • 每个核心方向一个Sheet
   • 保存路径：/Users/tututu/.openclaw/workspace/选股结果_YYYYMMDD_HHMMSS.xlsx

输出格式

文本回复

【核心利好方向分析】

根据对内容的分析，**聚焦**以下核心利好方向：

## 一、XX行业/概念（最核心）
- **利好逻辑**：具体说明为什么利好（原因+传导路径）
- **产业链位置**：上游/中游/下游
- **核心标的**：YYY、ZZZ

## 二、XX行业/概念（次核心）
- **利好逻辑**：...
- ...

---

### 选股结果

## 一、XX概念相关股票
| 序号 | 代码 | 名称 | 现价 | 涨跌幅 | 核心逻辑 |
|------|------|------|------|--------|----------|
| 1 | XXXXXX | XXXX | XX.XX | X.XX% | ... |
| 2 | XXXXXX | XXXX | XX.XX | X.XX% | ... |
| ... | ... | ... | ... | ... | ... |

## 二、YY概念相关股票
...（同上）

---
📊 Excel文件已生成: 选股结果_20260312_163000.xlsx

Excel文件结构

• 文件名: 选股结果_YYYYMMDD_HHMMSS.xlsx
• Sheet: 按核心方向名称命名（最多5个Sheet）
• 列: i问财返回的原始字段名
• 行: 每个方向最多5只股票

注意事项

1. 聚焦优先：只识别最核心的3-5个方向，不要列出10+个方向
2. 必须有逻辑：每个方向都要说明"利好逻辑"，不能只给结论
3. 产业链位置：标注清楚在产业链的哪个环节
4. 字段完整性：获取并显示所有返回字段
5. 异常处理：某查询失败时记录错误，继续其他查询