concept-research

concept-research is a skill for Claude Code from Howard-Jerry/quant-agent-skills. It costs 101 tokens per session (6,905 once invoked), scanned A, original, MIT.

A research workflow for tracking the life cycle of an investment theme and identifying the companies or other assets best connected to it. It separates broad industry research, theme-based research, and analysis of individual stocks.

In plain words
What is it for?
Use it to scan themes, assess catalysts and leaders, judge the theme’s stage, and hand a selected company to a separate stock-research process.
Why use it?
It helps distinguish a developing market story from an overheated or fading one before choosing what to investigate further.

Skill for Claude Code

Written for Claude Code: allowed-tools in frontmatter. Also seen: mentions CLAUDE.md.

Needs its repository: it runs a file that does not travel with it, so clone the repository first. The line is python scripts/concept_radar.py # 默认 top 20.

Good fit Use it to scan themes, assess catalysts and leaders, judge the theme’s stage, and hand a selected company to a separate stock-research process.

Compare 6 skills from other repositories ↓
Install

Getting it into your agent

It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.

Clone the repo
git clone --depth 1 https://github.com/Howard-Jerry/quant-agent-skills
agentmods
npx agentmods add skills/howard-jerry/quant-agent-skills/concept-research

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for concept-research

README.md
[![agentmods](https://agentmods.dev/badge/skills/howard-jerry/quant-agent-skills/concept-research/github.svg)](https://agentmods.dev/skills/howard-jerry/quant-agent-skills/concept-research)
Your own site
<a href="https://agentmods.dev/skills/howard-jerry/quant-agent-skills/concept-research"><img src="https://agentmods.dev/badge/skills/howard-jerry/quant-agent-skills/concept-research/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for concept-research

Your own site · 80×15
<a href="https://agentmods.dev/skills/howard-jerry/quant-agent-skills/concept-research"><img src="https://agentmods.dev/badge/skills/howard-jerry/quant-agent-skills/concept-research.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 101 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 6,905 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00101 $0.06905
Opus 5 $0.00051 $0.03453
Sonnet 5 $0.00020 $0.01381
Haiku 4.5 $0.00010 $0.00690

Measured 9d ago against content hash 4c9381743eb2, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

concept-research scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

concept-research/SKILL.md · 397 lines

How it starts

The opening of the file, as written. The whole thing — 397 lines — stays where its author put it; the contents beside it link to each section on GitHub.

公开版适配说明(v1.0.0):本 skill 来自作者个人量化系统,命令和文件路径是 作者技术栈的具体实现。使用前请先读术语表与适配指南: https://github.com/Howard-Jerry/quant-agent-skills/blob/main/docs/adaptation-guide.md {{QUANT_ROOT}} 是你的量化项目根目录;RRS/Vault/scripts/*.py 等是作者 配套组件,公开版不附带,请按指南替换为你自己的数据层与知识库(最小骨架见仓库 template/ 目录)。

concept-research:概念研究编排器 + 三地知识沉淀

你是量化概念研究编排器(Conductor)。核心职责:追踪市场叙事的生命周期,在故事的不同阶段找到最佳投资载体。

工作区隔离(建议,非强制):编辑 common/auto_research/ 代码时先 EnterWorktree。五文件输出可直接在 auto_research/knowledge/concept_research/ 下写。EnterWorktree 失败时 fallback 到主 repo 路径。

stock-research / industry-research 的关系:

  • industry-research = top-down 行业配置;
  • concept-research = narrative-centric 叙事生命周期到最佳载体;
  • stock-research = bottom-up 个股估值到交易计划;
  • 概念研究锁定最佳载体后,调用 stock-research 做单票 L2/L3 深挖。

核心哲学:叙事生命周期驱动的投资决策

概念投资决策 = f(叙事阶段, 龙头质量, 入场时机)

叙事生命周期:
  形成期 → 催化剂出现, 少数先知先觉, 概念指数微涨放量
  加速期 → 共识形成, 资金涌入, 龙头确立, 概念指数快速拉升
  扩散期 → 跟风蔓延, 散户入场, 真假龙混杂, 概念指数高位震荡
  退潮期 → 叙事疲劳, 资金流出, 龙头派发, 概念指数回落缩量

你的工作不是"猜哪个概念会火"——是搞清楚一个概念现在处于什么阶段,龙头是谁,这个阶段该怎么交易

  1. 这个故事为什么现在爆发?(催化剂是什么?能持续多久?)
  2. 龙头是谁?是真龙还是假龙?(有业绩兑现 vs 纯叙事炒作)
  3. 现在是好的介入时点吗?(叙事阶段 × 量价形态 × 资金面)
  4. 该怎么交易?(入场/止损/仓位/退出信号)

公告披露不是叙事加速的充分条件。任何公告、财报、订单、客户认证、产能投放、政策落地或产品发布,都必须先过 event-expectation-delta-guard:写清 pre_event_expectationpriced_in_viewactual_disclosuredelta_class=BEAT|IN_LINE|MISS|AMBIGUOUSaction_boundaryIN_LINE 是利好落地/继续观察,不得直接升级 L3、龙头排序、入场线、仓位或 trigger;BEAT 也只能触发进一步验证,不自动买入。

你的结构性优势

  • 可以跨行业追踪概念(机构受限于行业分工,看得到自己那一亩三分地)
  • 可以等最佳时点(机构有排名压力,不得不追涨)
  • 可以分辨真假龙(机构持仓限制不能碰小票,但真龙往往不在成分股最大市值里)
  • 只对自己诚实(机构有职业风险——"跟着概念一起错比一个人错安全")

概念簇原则:叙事不是总等于单一 concept_code

市场叙事常常横跨多个同花顺概念。比如"半导体材料/设备国产替代"不能只看"光刻机"或"光刻胶",而要同时看光刻胶、光刻机、中芯国际概念、国家大基金持股等概念的并集/交集。

遇到以下情况,L2 必须按概念簇处理:

  • L1 Top20 中多个概念属于同一叙事链;
  • 新闻/研报催化明确跨材料、设备、客户、政策资金等多条子线;
  • 龙头候选在多个概念里重复出现;
  • 单一概念指数强,但成分股解释不了真实资金方向。

概念簇 L2 必做:

  1. 列出覆盖的 concept_code -> concept_name
  2. 计算成分股并集、交集、concept_count
  3. 分别报告每个概念指数的日期、5/20/60日涨幅、120日位置、量比;
  4. 用最近完整个股截面计算并集扩散度:上涨比例、>5%比例、5日/20日中位数;
  5. 龙头排序同时看涨幅、概念重叠数、总市值/流动性、财务与资金面;
  6. 结论必须说明"单点概念"还是"概念簇",避免把叙事切碎或重复研究。

Read the full file on GitHub · 397 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 9d ago First seen · 397 lines · 101 tokens per session scan A 4c9381743eb2

Subscribe to this mod's changes

concept-research is a skill published in the GitHub repository Howard-Jerry/quant-agent-skills (2 stars, last pushed 1mo ago), licensed MIT. It adds 101 tokens to every session and 6,905 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other skills, from other repositories

grayscale-crypto-sectors

Use when evaluating crypto through a Grayscale-style Crypto Sectors lens: sector taxonomy, FTSE/Grayscale index eligibility, fee/usage fundamentals, sector-share valuation, ETP/trust wrappers, and Zcash-style privacy-as-money theses.

questflowai/investorskills · 56 tokens

ansem-crypto

Use when evaluating crypto narratives, attention rotation, memecoin cycles, Solana-style ecosystem momentum, social distribution, and reflexive retail flows in an Ansem-style crypto market framework.

questflowai/investorskills · 41 tokens

serenity

Use when evaluating US stocks through Serenity / @aleabitoreddit's AI and semiconductor supply-chain lens: upstream chokepoints, photonics/CPO bottlenecks, hyperscaler capex, institutional rotation, dilution risk, and thesis validation.

questflowai/investorskills · 52 tokens

tqsdk-trading-and-data

Explain, implement, or debug TqSdk Python workflows for waitupdate or ischanging update loops, market data retrieval, historical download, account type selection, funds or positions or orders or trades, field meanings, order placement or cancellation, target-position tools, TqScenario margin trials, real-account…

shinnytech/tqsdk-python · 286 tokens

buffett

Use when evaluating a business through Buffett-style ownership, owner earnings, durable moat, management quality, capital allocation, and margin-of-safety judgment.

questflowai/investorskills · 32 tokens

druckenmiller

Use when evaluating top-down macro trades, liquidity regimes, currency/rates/equity themes, concentrated risk-taking, and quick loss-cutting in a Stanley Druckenmiller style.

questflowai/investorskills · 42 tokens