wb-finance-skill

wb-finance-skill is a skill for Claude Code from rojim666/SztuCode. It costs 118 tokens per session (7,583 once invoked), scanned A, original, MIT.

A finance-focused instruction set for requests about markets, investments, companies, funds, trading, and related financial topics. It requires checking current data through a search tool and following rules against invented figures and unsupported advice.

In plain words
What is it for?
Use it to route finance questions, retrieve market information, distinguish financial terms correctly, handle conflicting sources, and format investment-related answers with the required warning.
Why use it?
Financial information can become outdated or be confused across similar concepts. These instructions require source checking, consistency checks, and a standard disclaimer for specific investment recommendations.

Skill for Claude Code

Written for Claude Code: when-to-use in frontmatter.

Good fit Use it to route finance questions, retrieve market information, distinguish financial terms correctly, handle conflicting sources, and format investment-related answers with the required warning.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/rojim666/sztucode/wb-finance-skill
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add rojim666/SztuCode --skill wb-finance-skill
Clone the repo
git clone --depth 1 https://github.com/rojim666/SztuCode

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 wb-finance-skill

README.md
[![agentmods](https://agentmods.dev/badge/skills/rojim666/sztucode/wb-finance-skill/github.svg)](https://agentmods.dev/skills/rojim666/sztucode/wb-finance-skill)
Your own site
<a href="https://agentmods.dev/skills/rojim666/sztucode/wb-finance-skill"><img src="https://agentmods.dev/badge/skills/rojim666/sztucode/wb-finance-skill/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 wb-finance-skill

Your own site · 80×15
<a href="https://agentmods.dev/skills/rojim666/sztucode/wb-finance-skill"><img src="https://agentmods.dev/badge/skills/rojim666/sztucode/wb-finance-skill.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 118 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 7,583 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 $0.00118 $0.07583
Opus 5 $0.00059 $0.03792
Sonnet 5 $0.00024 $0.01517
Haiku 4.5 $0.00012 $0.00758

Measured today against content hash a0441f732c75, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

wb-finance-skill 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 today.

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.

packages/runtime-ts/prompts/workbuddy/skills/wb-finance-skill/SKILL.md · 223 lines

How it starts

The opening of the file, as written. The whole thing — 223 lines — stays where its author put it; the contents beside it link to each section on GitHub.

触发顺序硬约束

  • 必须加载 wb-finance-skill 读取红线、时区口径、路由与输出规范,任何涉及金融市场数据的请求,都要调用 agentic_search 工具检索数据;禁止跳过本 skill 直接裸答或者凭记忆回答。

红线(金融场景一票否决)

  • 禁止编造数据:不虚构数据/事件/公司名/财务数字;数据源缺失时直接说明"当前数据源未覆盖 / 需进一步核实",不要编一组数据再加"待核实"标签;引用不确定的研报/论文时标"该引用需核实原文"

  • 禁止核心概念混淆:客户 vs 竞争对手、整机厂 vs 零部件厂、净利润 vs 归母净利润、同比 vs 环比、财年 vs 自然年;不确定时用"据我理解"前缀并请用户确认

  • 禁止数据自相矛盾:同一回答内数据与结论必须一致;多组数据先交叉校验;数据源冲突时优先采信高层级来源(交易所、公司公告、年报)并显式标注分歧

  • 强制免责声明:所有包含具体投资建议、操作价位、买卖判断、仓位调整建议的输出,必须在回复末尾附加以下免责声明模板(固定文案,禁止模型自行改写、缩减或省略):

    免责声明:以上内容基于公开数据和量化分析,仅供参考,不构成投资建议。市场有风险,投资需谨慎。任何投资决策应结合个人风险承受能力、资金状况和投资目标独立判断,必要时咨询持牌专业机构。过往表现不预示未来收益。

检索策略

  • 金融数据检索统一经 agentic_search 工具 —— 它具备自主分析与多步规划能力,会自行判断查哪些维度、分几步查,返回结论。
  • 委派 query 必须是"一句话检索意图",保留用户原始意图,禁止拆成多维度清单 / 字段列表 / 表格格式要求。工具 自身具备多步规划与自主检索能力,会自行拆维度、判断查哪些字段、查多深——你拆得越细、要求越"全",它解锁的检索面越大、越发散、越慢。委派时只交代两件事:① 标的 / 主题 / 范围(带代码),② 大方向查什么;其余(查哪些字段、列几列、怎么排序、要不要表格、分几个维度)一律不写,也不要要求工具 写分析 / 结论报告 / 大段表格——它只需返回结论。字段筛选、表格化、排序、深度分析都是拿回数据后主 agent 自己的活(见第 4/7 条),不是委派 query 的内容。
  • 不要要求"全面/详细/深入"检索:委派里禁止出现"请尽可能全面地检索""详细检索""返回结构化分析数据""覆盖以下 N 个方面"这类堆砌词。检索广度与深浅由工具 按问题体量自己定,主 agent 说得越"全""细"它越发散、越慢,反而不利。就给它一句朴素的检索意图即可。
    • 反例(过度拆解,禁止):用户问"列出场内基金里红利低波和红利自由现金流 ETF",却委派"请查询 A 股场内 ETF 中红利低波、红利自由现金流两主题的所有相关 ETF,列出基金代码、简称、跟踪指数、管理人、最新规模、近一周/近一月/年初至今涨跌幅、管理费率+托管费率、成立日期,用表格分主题输出……"
    • 正例(一句话):"列出场内(A 股)红利低波、红利自由现金流两个主题的相关 ETF"
    • 反例:用户问"国内有哪些上市公司跟 SpaceX 相关",却委派"请全面检索 A 股与 SpaceX 有业务关联的公司,覆盖直接供应商 / 产业链相关 / 对标概念 / 最新动态四个维度,列出名称、代码、关联逻辑、近期表现,区分实际业务与概念炒作……"
    • 正例:"检索 A 股里与 SpaceX 相关的上市公司"
    • 反例(宽泛研究被拆维度 + 堆"全面"词,禁止):用户问"银河电子怎么样",却委派"全面分析 A 股银河电子(002519),需覆盖:1 主营业务 2 财务数据 3 估值 4 股价走势与资金流向 5 研报评级 6 近期公告 7 板块概念,请尽可能全面地检索并返回结构化分析数据"
    • 正例:"研究 A 股银河电子(002519)的整体情况"
  • 宽泛的"X 怎么样 / 值不值得看"只需一句话点明标的与代码,具体查什么、查多全由工具 自己拆;只有用户问题本身就很具体、只问单一字段时(如"茅台最新 PE")才如实精确转述。
  • 工具无法满足时,用 WebSearch 检索公开信息,明确告知用户数据来源并说明非实时性。

数据底线

  • 前提显式:问操作类问题(买/卖/加仓/减仓/换股)时,先列前提(市场环境 + 用户风险偏好 + 资金量/期限),再给"条件 → 操作 → 风险提示"。前提缺失时主动追问而非直接给操作建议
  • 检索优先于记忆:提及具体股票/基金/指数/宏观指标时,先调 agentic_search;如通达信 MCP 可用,按"数据查询优先级策略"中的场景调用拉数据,禁止纯凭记忆作答;记忆中的数字只能作为合理性 sanity check,不能作为答案
  • 禁止硬编码数据:所有行情、财务、宏观和技术指标必须通过工具动态获取并标注来源和时点,禁止在回答中直接引用训练数据中的历史数值或凭记忆输出数字
  • 时效意图与目标周期解析:用户表达“最新、当前、今天、今年、近期”等时效要求时,先结合运行时日期、市场交易状态、指标发布频率和数据发布时间确定目标周期,不得把当前年份直接等同于最新有效数据周期。用户明确指定历史日期、年份、季度、财年或回测时点时,以用户指定范围为准,不得自动改写为当前周期。

Read the full file on GitHub · 223 lines

Files

What ships with it

46 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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. today First seen · 223 lines · 118 tokens per session scan A a0441f732c75

Subscribe to this mod's changes

wb-finance-skill is a skill published in the GitHub repository rojim666/SztuCode (64 stars, last pushed today), licensed MIT. It adds 118 tokens to every session and 7,583 once invoked, about $0.0006 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 2026-09-12.