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.
npx skills add vivy-yi/finance-skills --skill tax-risk-assessmentgit clone --depth 1 https://github.com/vivy-yi/finance-skillsWrote 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.
[](https://agentmods.dev/skills/vivy-yi/finance-skills/tax-risk-assessment)<a href="https://agentmods.dev/skills/vivy-yi/finance-skills/tax-risk-assessment"><img src="https://agentmods.dev/badge/skills/vivy-yi/finance-skills/tax-risk-assessment/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.
<a href="https://agentmods.dev/skills/vivy-yi/finance-skills/tax-risk-assessment"><img src="https://agentmods.dev/badge/skills/vivy-yi/finance-skills/tax-risk-assessment.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00104 | $0.02062 |
| Opus 5 | $0.00052 | $0.01031 |
| Sonnet 5 | $0.00021 | $0.00412 |
| Haiku 4.5 | $0.00010 | $0.00206 |
Grade A, and why
tax-risk-assessment 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.
How it starts
The opening of the file, as written. The whole thing — 256 lines — stays where its author put it; the contents beside it link to each section on GitHub.
加载上下文
首次使用时： 读取 ../../CLAUDE.md 获取场景级配置（税负率参考/风险等级/升级矩阵）。
/tax-risk-assessment — 税务风险自检
Examples
→ 示例：用户说"帮我们评估一下目前的税务风险，主要关注哪些点"，系统应调用本技能，执行税务风险评估。
→ 示例：用户说"新业务涉及跨境服务，帮我看看有没有新的税务风险"，系统应调用本技能，评估跨境业务的税务风险。
→ 示例：用户说"税局最近查得比较严，帮我看看我们的合规性有没有问题"，系统应调用本技能，评估当前税务合规性状态。
风险来源
风险来源一：金税系统预警（[TAX-GOLD]）
→ 系统自动监控，发现异常行为
风险来源二：申报数据异常
→ 税负率/进销项/收入成本配比异常
风险来源三：发票风险
→ 失控发票/异常凭证/虚开发票
风险来源四：外部信息
→ 协查函/举报/审计署/税务局评估通知
第一步：从金税系统获取预警
查询路径（[TAX-GOLD]）：
登录金税系统纳税人端
→ 风险监控 → 风险预警处理
→ 查看所有橙色预警和红色预警
→ 逐项查看预警详情
常见预警类型：
增值税预警：
- 税负率预警：实际税负率低于行业参考值
- 进销项异常：进项长期大于销项
- 农产品收购发票异常：收购金额与产能不匹配
- 旅客运输发票异常：抵扣金额与业务不匹配
- 发票顶额开具：单张发票开具面额接近最高限
企业所得税预警：
- 收入成本配比异常：成本占比异常高/低
- 期间费用率异常：费用占比异常
- 毛利率异常：与行业均值差异大
其他预警：
- 纳税信用等级下降
- 异常协查函
第二步：逐项核实预警
核实流程：
对每个预警项：
第一步：确认数据准确性
→ 预警数据与实际数据是否一致
→ 如不一致：预警数据可能有误
第二步：判断是否为合理商业原因
→ 有合理原因：准备书面说明
→ 无合理原因：确认为真实风险
第三步：评估风险等级
→ 红色预警：高风险，须立即处理
→ 橙色预警：中风险，须关注跟进
→ 黄色预警：低风险，持续观察
常见预警的合理原因：
税负率偏低：
合理原因：
- 业务季节性波动（淡旺季）
- 存货结构变化（期初库存大/期末库存小）
- 采购集中（集中采购导致进项多）
- 产品结构变化（低税率产品占比增加）
不合理原因：
- 隐匿销售收入
- 虚增进项税额
进项长期大于销项：
合理原因：
- 业务扩张期（存货储备增加）
- 准备过节/促销（提前采购）
不合理原因：
- 取得虚开发票
- 滞留票未及时认证
第三步：发票风险核查
失控发票核查：
□ 是否收到过失控发票？
□ 如收到：是否已抵扣？是否已做进项转出？
□ 如已抵扣：须向税务机关申请核实
□ 是否向供应商追索正规发票？
□ 是否保留了相关证据（合同/物流单据/付款凭证）？
异常凭证核查：
□ 是否有发票被认定为异常凭证？
□ 如有：是否在 60 日内向税务机关申请核实？
□ 异常原因是什么？能否提供证明材料？
上下游风险传导：
□ 上游供应商是否存在税务问题？
□ 如存在：须评估对我公司的影响
□ 是否须做进项转出？
□ 是否须向税务机关主动报告？
第四步：申报数据交叉检查
增值税 vs 企业所得税：
□ 增值税申报收入与所得税申报收入是否一致？
□ 差异原因是什么？（暂时性差异/永久性差异/数据错误）
□ 如差异 > 10%：须查明原因
发票数据 vs 申报数据：
□ 开票系统销项合计 vs 增值税申报表销项
□ 勾选平台进项合计 vs 增值税申报表进项
□ 如不一致：须在申报期内更正申报
第五步：生成风险报告
═══════════════════════════════════════
税务风险评估报告
评估期间：[YYYY年MM月]
评估范围：[全税种/增值税/企业所得税/个人所得税]
生成时间：[YYYY-MM-DD HH:MM]
═══════════════════════════════════════
【金税系统预警概览】
红色预警（高风险）：[X] 项
橙色预警（中风险）：[X] 项
黄色预警（低风险）：[X] 项
【重点预警详情】
[红色预警 1]
预警类型：[税负率偏低/进销项异常/...]
触发条件：[具体描述]
涉及金额：[X万元]
核实结果：
[ ] 误报（有合理商业原因）
[ ] 真实风险
原因分析：[...]
整改措施：[...]
整改期限：[YYYY-MM-DD]
负责人：[姓名]
[橙色预警 1]
...
【发票风险】
失控发票：[X] 张，已抵扣 [X] 张
→ 已转出进项：[X万元]
→ 追索正规发票：[X] 张
异常凭证：[X] 张
→ 已在 60 日内申请核实：[✅/❌]
上下游风险：[描述]
影响评估：[...]
【申报数据交叉检查】
增值税 vs 所得税收入差异：[X%] [✅ 一致 / ⚠️ 有差异]
差异原因：[...]
开票 vs 申报销项差异：[✅ 一致 / ⚠️ 有差异]
开票 vs 申报进项差异：[✅ 一致 / ⚠️ 有差异]
【整改计划】
优先级 | 风险项 | 整改措施 | 整改期限 | 负责人
--------|--------|---------|---------|--------
P0 | [风险] | [措施] | [日期] | [姓名]
P1 | [风险] | [措施] | [日期] | [姓名]
...
【纳税信用】
当前信用等级：[A级/B级/M级/C级/D级]
较上期变化：[提升/下降/无变化]
原因：[...]
═══════════════════════════════════════
整体风险等级：[🟢 低风险 / 🟡 中风险 / 🔴 高风险]
须升级至税务经理：[✅ 是 / ❌ 否]
═══════════════════════════════════════
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.
- 9d ago First seen · 256 lines · 104 tokens per session scan A ef7959ef78e8
tax-risk-assessment is a skill published in the GitHub repository vivy-yi/finance-skills (29 stars, last pushed 3mo ago), licensed MIT. It adds 104 tokens to every session and 2,062 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 2026-09-03.
Other skills, from other repositories
iso-13485-certification
Comprehensive toolkit for preparing ISO 13485 certification documentation for medical device Quality Management Systems. Use when users need help with ISO 13485 QMS documentation, including (1) conducting gap analysis of existing documentation, (2) creating Quality Manuals, (3) developing required procedures and work…
fda-database
Query openFDA API for drugs, devices, adverse events, recalls, regulatory submissions (510k, PMA), substance identification (UNII), for FDA regulatory data analysis and safety research.
see-invisibility
In D&D, See Invisibility lets you see creatures and objects that have been made invisible. The real-world version is revealing deliberate obscurity: finding the hidden costs in a pricing page, uncovering the actual terms buried in a EULA, identifying the obfuscated tracking in a codebase, or surfacing the real…
dietary-check
Audit a menu for allergen labelling, vegan and halal coverage, and common dietary requirements.
patentforge-provisional
Use this skill when the user wants to turn an invention idea into a structured invention disclosure, prior art search strategy, claim candidates, and provisional patent draft package. This skill is for patent drafting assistance only and must not provide legal advice or guarantee patentability.
healthcare-marketing-compliance
Skill "healthcare-marketing-compliance" from aimerdoux/wavex-os, covering healthcare marketing compliance specialist, your identity & memory, core mission, medical advertising compliance and pharmaceutical marketing standards.