A
AbhijeetBabar-2025/stock-analyst-skill
Stock analyst skill for Claude Code & AI agents: quick scores, full reports, portfolio tracking, and a…
无 SECURITY.md
在 GitHub（仓库已不存在）链接已于 2026-11-12 核查 · 自动访问被拒（HTTP 429），状态未能确认
安装与配置
客户端配置（取自该仓库的安装命令）
{
"mcpServers": {
"abhijeetbabar-2025-stock-analyst-skill": {
"command": "npx",
"args": ["skills"]
}
}
}解析自我们于 2026-10-26 抓取的该仓库 README；我们未执行过。
command/args 来自仓库写明的安装命令；不确定时请以该项目 README 为准。
VS Code（.vscode/mcp.json —— 键名是 servers）
{
"servers": {
"abhijeetbabar-2025-stock-analyst-skill": {
"command": "npx",
"args": ["skills"]
}
}
}VS Code 顶层键是 <code>servers</code>，其余与上面那份相同。
Claude Code（命令行）
claude mcp add --transport stdio abhijeetbabar-2025-stock-analyst-skill -- npx skills
Claude Code 不收 JSON 文件，用这条命令直接加。
配置项的键名以你的客户端为准：客户端与 CLI
- ★ 数取值
- 0 ★ · AbhijeetBabar-2025/stock-analyst-skill · 取自 GitHub 搜索 API · （2026-10-26 20:00:26 核对）
- 来源明细
- 从该仓库的 GitHub 页面直接读取 · 条目类型：开源项目
- 收录与活跃
- 收录于 2026-10-23 · 最近见到 2026-11-14 · 共见到 26 次 · 被 1 个来源列出
- 评分构成
- 综合分 103 · 1 个来源 · repo
- 安装
npx skills
常见问题（答案都来自我们的记录）
怎么安装？
该仓库给出的安装命令是
npx skills。我们的记录里它是仓库自己提供的；不保证在你的客户端上直接可用。它适合谁 / 用来做什么？
我们把它归在这些标签下: Agent 技能。标签来自条目自身，不是我们强加的分类。
谁维护它、还在更新吗？
仓库
AbhijeetBabar-2025/stock-analyst-skill；我们收录于 2026-10-23；最近见到 2026-11-14。这里没有"官方认证"或"质量评级"——我们只记录来源与时间。# Stock Analyst skill
以下摘录自该仓库的 README（片段） · 在 GitHub 读完整 README →
A stock analyst skill for Claude Code and other AI agents. It learns how you invest and improves from its own track record.
📖 Beginner's installation and user guide
🎬 Demo (21 seconds; click to watch with sound)
Quick score: a -100 to +100 score for any Yahoo Finance ticker (US, India .NS/.BO, London .L, crypto -USD). It combines 7 factors: valuation, growth, profitability, financial health, momentum, analyst views and earnings surprises.
Full report: fundamental and technical research with bull and bear cases, valuation and risks, using live web research.
ETFs: funds (GLD, SPY, TLT, NIFTYBEES.NS) get their own score from momentum, track record, risk, cost and liquidity, and are compared with the right index for their asset class.
Dashboard: a local web page to type in symbols, see each result as a card, save a watchlist and track past calls. Click a card for the full analysis: why it's moving, latest headlines, bull and bear cases, what to watch, cautions, research gaps, learning-only trading notes and Claude's saved report. A Portfolio tab shows holdings with totals per currency, allocation, profit/loss and today's score. Run uv run ~/.claude/skills/stock-analyst/scripts/dashboard.py --open.
Comparison: puts several tickers side by side and ranks them.
Portfolio tracker: your holdings with profit/loss and warnings when one holding is too large.
Memory that improves over time:
Every call goes into a journal.
After 30 days, review checks each call against what the stock and its market index actually did, and turns the misses into lessons.
The skill reads those lessons and your investor profile before every new analysis.
Factor weights can be re-tuned from the evidence, but only with your approval.
## Install
bash
npx skills add AbhijeetBabar-2025/stock-analyst-skill
Add -g to install it for your user instead of the current project.
Already installed? Update with npx skills update stock-analyst -g -y, then restart Claude Code (and the dashboard, if it's running). Your data is kept. See What's new.
Requirements: uv, used to run the Python scripts. It installs yfinance and pandas automatically on first run. No API keys are needed.
## Use
Just ask your agent:
"Quick score on NVDA and AMD"
"Full report on RELIANCE.NS"
"Compare MSFT vs GOOGL"
"I bought 10 AAPL at 180"
"Show my portfolio"
"Review how our past calls did"
You can also run the scripts directly:
bash
S=~/.claude/skills/stock-analyst # or wherever the skill was installed
uv run $S/scripts/score.py AAPL TCS.NS BTC-USD
uv run $S/scripts/portfolio.py show
uv run $S/scripts/journal.py review
## Your data
Everything personal stays on your machine in ~/.stock-analyst/ (change the location with STOCKANALYSTDATA). Skill updates never touch it.
| File | What it holds |
| profile.md | How you invest, filled in as it comes up |
| lessons.md | What reviews and your corrections taught it |
| journal.jsonl | Every call: date, price, verdict, reason |
| portfolio.json | Your holdings |
| watchlist.json | Symbols saved in the dashboard |
| reports/ | Full reports, shown in the dashboard |
📖 Beginner's installation and user guide
🎬 Demo (21 seconds; click to watch with sound)
Quick score: a -100 to +100 score for any Yahoo Finance ticker (US, India .NS/.BO, London .L, crypto -USD). It combines 7 factors: valuation, growth, profitability, financial health, momentum, analyst views and earnings surprises.
Full report: fundamental and technical research with bull and bear cases, valuation and risks, using live web research.
ETFs: funds (GLD, SPY, TLT, NIFTYBEES.NS) get their own score from momentum, track record, risk, cost and liquidity, and are compared with the right index for their asset class.
Dashboard: a local web page to type in symbols, see each result as a card, save a watchlist and track past calls. Click a card for the full analysis: why it's moving, latest headlines, bull and bear cases, what to watch, cautions, research gaps, learning-only trading notes and Claude's saved report. A Portfolio tab shows holdings with totals per currency, allocation, profit/loss and today's score. Run uv run ~/.claude/skills/stock-analyst/scripts/dashboard.py --open.
Comparison: puts several tickers side by side and ranks them.
Portfolio tracker: your holdings with profit/loss and warnings when one holding is too large.
Memory that improves over time:
Every call goes into a journal.
After 30 days, review checks each call against what the stock and its market index actually did, and turns the misses into lessons.
The skill reads those lessons and your investor profile before every new analysis.
Factor weights can be re-tuned from the evidence, but only with your approval.
## Install
bash
npx skills add AbhijeetBabar-2025/stock-analyst-skill
Add -g to install it for your user instead of the current project.
Already installed? Update with npx skills update stock-analyst -g -y, then restart Claude Code (and the dashboard, if it's running). Your data is kept. See What's new.
Requirements: uv, used to run the Python scripts. It installs yfinance and pandas automatically on first run. No API keys are needed.
## Use
Just ask your agent:
"Quick score on NVDA and AMD"
"Full report on RELIANCE.NS"
"Compare MSFT vs GOOGL"
"I bought 10 AAPL at 180"
"Show my portfolio"
"Review how our past calls did"
You can also run the scripts directly:
bash
S=~/.claude/skills/stock-analyst # or wherever the skill was installed
uv run $S/scripts/score.py AAPL TCS.NS BTC-USD
uv run $S/scripts/portfolio.py show
uv run $S/scripts/journal.py review
## Your data
Everything personal stays on your machine in ~/.stock-analyst/ (change the location with STOCKANALYSTDATA). Skill updates never touch it.
| File | What it holds |
| profile.md | How you invest, filled in as it comes up |
| lessons.md | What reviews and your corrections taught it |
| journal.jsonl | Every call: date, price, verdict, reason |
| portfolio.json | Your holdings |
| watchlist.json | Symbols saved in the dashboard |
| reports/ | Full reports, shown in the dashboard |
收录徽章（可选）
想把收录标记放进 README 就复制这行，可选，不影响收录。
[](https://anclum.com/zh/s/abhijeetbabar-2025-stock-analyst-skill/)
- 来源
- github · https://github.com/AbhijeetBabar-2025/stock-analyst-skill
- 按领域浏览
- Agent 技能
- 收录于
- 今天
- 安全策略
- 我们没在该仓库根目录找到
SECURITY.md（这不代表不安全）
同类推荐
所属领域与标签
每个标签都能点进去看同类的全部条目。数字是当前库里的真实条数。