Skills Explorer Add Skills

daymade-sector-research

An agent skill by daymade, from daymade/claude-code-skills. Tags: analytics, market-analysis, multi-agent, research, trading.

What it does

A股行业投研工作流：全板块 Top N 涨幅标的计算、公告窗口检索（周/月）、市场情绪证据分级判断，并以 Agent Team 并行执行 + fresh-context 对抗验证。 触发场景（命中任一即读本 skill）： - 「XX 行业 Top N / Top 10 标的」（如「医药行业今天的 Top 10 标的」） - 「这些标的最近一周 / 一个月发过哪些公告」 - 「判断 XX 行业 / 板块现在的市场情绪怎么样」 - 「用 Agent Team / 多 agent 做行业投研」或要求对行业研究结论做对抗性验证 - 需要「基于证据、基于数据」的板块级研究交付（成分股 → 行情 → 公告 → 情绪 → 验证 → 报告） 核心能力：东财 push2 成分股 → 新浪实时快照涨幅排序（scripts/top_n_pipeline.py）；巨潮 cninfo + 东财公告双源交叉，沪深+北交所统一通道（scripts/ann_query.py）；情绪证据 L1（一手行情）/L2（带时间戳媒体）/L3（未核实标题）分级；Agent Team 并行编排与验证纪律。全程国内公开数据源，无付费依赖；Gangtise 官方 skill 为可选增强（需积分，缺失时走公开源 pivot）。 最高纪律（用户原话）：没有十足把握的事，宁可标注不确定，不可给错误答案。

Install

With the skills CLI, which installs into Claude Code, Codex, Cursor and other agents:

npx skills add daymade/claude-code-skills --skill daymade-sector-research

Or copy the skill folder into Claude Code's skills directory by hand (~/.claude/skills for every project, or .claude/skills inside one):

git clone --depth 1 https://github.com/daymade/claude-code-skills
cp -r claude-code-skills/daymade-financial/daymade-sector-research ~/.claude/skills/daymade-sector-research

Safety box score

Grade B, low risk: 21 of 100 risk points, rated 2026-10-20. The grade rates what the skill and its scripts can reach on the machine of whoever installs it, across eight categories from shell execution to secrets access. How the score works.

This skill is a live A-share sector-research pipeline: it fetches market and announcement data from Eastmoney, Sina, and cninfo, then writes several CSVs and orchestrates multi-agent verification. The main risks are ordinary internet access plus semi-autonomous agent coordination; it does not request credentials or elevated privileges.

Category levels, from 0 (none) to 3 (high)
CategoryLevelWhy
Code & shell execution1 (low)It ships runnable Python (`scripts/top_n_pipeline.py`, `scripts/ann_query.py`) that executes live HTTP-fetching and CSV-writing code, but it does not eval downloaded code or require shelling out beyond standard-library I/O.
Network access2 (moderate)The workflow depends on external endpoints (`push2.eastmoney.com`, `hq.sinajs.cn`, `cninfo.com.cn`, `np-anotice-stock.eastmoney.com`) and even suggests direct-IP access via `curl --resolve`, so installing it grants broad outbound data access.
Filesystem reach1 (low)The scripts create timestamped CSVs like `universe_...csv`, `top...csv`, `all_quotes...csv`, `distribution...csv`, and `announcements_...csv` in a caller-chosen output directory; no wider filesystem access is specified.
Secrets & credentials0 (none)There is no request for API keys, tokens, passwords, `.env` files, keychains, or browser sessions.
Privilege & persistence0 (none)No sudo, global installs, service registration, scheduled tasks, or OS/shell configuration changes are required.
Instruction hijack surface0 (none)The files contain workflow rules and source citations, but no instructions to ignore safeguards, follow remote instructions, or hide actions from the user.
Irreversible actions2 (moderate)The skill explicitly uses multi-agent execution and `SendMessage` handoffs (`'完成后必须调 SendMessage...'`), so it can act semi-autonomously inside the agent runtime.
Transparency0 (none)All shipped files are readable Markdown/Python; there are no binaries, minified bundles, or decoded-at-runtime blobs.

Before installing

Rated 2026-10-20 by an automated review of 8 files from the skill's repository.

Source

Repository
daymade/claude-code-skills (all skills from this repository)
Path
daymade-financial/daymade-sector-research/SKILL.md
Branch
main
Updated
2026-09-20