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 guoliang1114-boop/AriaAI --skill data-analytics-anomaly-detectiongit clone --depth 1 https://github.com/guoliang1114-boop/AriaAIWrote 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/guoliang1114-boop/ariaai/data-analytics-anomaly-detection)<a href="https://agentmods.dev/skills/guoliang1114-boop/ariaai/data-analytics-anomaly-detection"><img src="https://agentmods.dev/badge/skills/guoliang1114-boop/ariaai/data-analytics-anomaly-detection/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/guoliang1114-boop/ariaai/data-analytics-anomaly-detection"><img src="https://agentmods.dev/badge/skills/guoliang1114-boop/ariaai/data-analytics-anomaly-detection.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00041 | $0.05482 |
| Opus 5 | $0.00020 | $0.02741 |
| Sonnet 5 | $0.00008 | $0.01096 |
| Haiku 4.5 | $0.00004 | $0.00548 |
Grade A, and why
data-analytics-anomaly-detection 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 12d 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 — 460 lines — stays where its author put it; the contents beside it link to each section on GitHub.
数据分析与异常检测
When To Use
- 审计中需要对大额交易数据集执行分析性程序时
- 需要识别潜在舞弊风险或异常交易时
- 客户数据量大，传统抽样方法可能遗漏异常时
- 需要评估财务数据的合理性和一致性时
- 关联方交易筛查和利益冲突识别时
- 持续审计或监控审计中需要自动化异常检测时
Tools
bash- 运行Python/R数据分析脚本write- 创建分析脚本和发现报告read- 读取交易数据文件（CSV、Excel）grep- 搜索数据文件中的特定模式
Framework
1. 本福特定律 (Benford's Law)
参照 Nigrini, M.J. (2012) "Benford's Law: Applications for Forensic Accounting, Auditing, and Fraud Detection"
本特定律预测自然数据集中首位数字的频率分布:
| 首位数字 | 预期频率 | 预期比例 |
|---|---|---|
| 1 | 30.1% | 0.301 |
| 2 | 17.6% | 0.176 |
| 3 | 12.5% | 0.125 |
| 4 | 9.7% | 0.097 |
| 5 | 7.9% | 0.079 |
| 6 | 6.7% | 0.067 |
| 7 | 5.8% | 0.058 |
| 8 | 5.1% | 0.051 |
| 9 | 4.6% | 0.046 |
适用条件:
- 数据集规模 ≥ 500条记录（理想 ≥ 1,000条）
- 数据跨越多个数量级（如1-10,000范围）
- 数据为自然生成而非人为设定（如价格表不适用）
- 不适用于受限数据（如工资在特定范围内）
统计检验方法:
- Z值检验: Z = (观测比例 - 预期比例) / 标准误差
- |Z| > 1.96 表示在95%置信水平下显著偏离
- |Z| > 2.58 表示在99%置信水平下显著偏离
- Chi-Square检验: χ² = Σ[(观测频次-预期频次)²/预期频次]
- df=8, χ²临界值(0.05) = 15.51
- MAD (Mean Absolute Deviation):
- MAD ≤ 0.006: 接近一致性
- 0.006 < MAD ≤ 0.012: 可接受
- 0.012 < MAD ≤ 0.015: 边缘可接受
- MAD > 0.015: 不可接受
2. 趋势分析 (Trend Analysis)
参照 ISA 520 Analytical Procedures 及 AICPA AU-C 520:
水平分析 (Horizontal Analysis):
- 同比分析 (YoY): (本期 - 上期) / 上期 × 100%
- 环比分析 (MoM/QoQ): 连续期间变动分析
- 异常阈值: 变动超过 ±10% 或 ±2个标准差需调查
垂直分析 (Vertical Analysis):
- 各科目占收入/总资产的百分比
- 与行业基准对比（参考公开行业数据）
- 结构异常阈值: 占比变动超过 ±3个百分点
趋势断裂检测:
- 移动平均偏离: 实际值偏离3期/6期移动平均超过2σ
- 线性回归残差: 残差超过 ±2σ的观测点
- 季节性调整后异常: 剔除季节性因素后的异常波动
关键比率分析:
- 毛利率波动: 与行业均值偏离超过 ±5个百分点
- 应收账款周转天数: 与行业均值偏离超过 ±15天
- 存货周转天数: 与行业均值偏离超过 ±20天
- 关联方往来余额占比异常增长
3. 重复检测 (Duplicate Detection)
完全重复检测:
- 金额+日期+供应商完全一致的发票
- 金额+日期+摘要完全一致的日记账分录
- 金额+银行账号完全一致的付款记录
近似重复检测:
- 金额相同(±0.01) + 日期相近(±3天) + 供应商不同
- 金额相同 + 供应商相同 + 发票号相似（编辑距离≤2）
- 反向分录: 同一科目金额相同方向相反，间隔≤30天
阈值设置:
- 完全匹配: 100%精确匹配
- 近似匹配: 金额容差±0.01，日期容差±3天
- 金额舍入: 金额为整百/整千的交易比例（正常应<10%）
4. 关联方筛查 (Related Party Screening)
参照 ISA 550 Related Parties 及 SEC Regulation S-K Item 404:
What ships with it
2 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.
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.
- 12d ago First seen · 460 lines · 41 tokens per session scan A d111e15e2baa
data-analytics-anomaly-detection is a skill published in the GitHub repository guoliang1114-boop/AriaAI (37 stars, last pushed today), licensed MIT. It adds 41 tokens to every session and 5,482 once invoked, about $0.0002 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 2026-08-30.
Other skills, from other repositories
company-brain
How an agent should operate as one mind inside a shared Company Brain built on Caura — recall before acting, obey fleet keystones, reuse and publish skills, and report outcomes so every task compounds across the whole organization. Use this whenever you work as part of a Caura-connected team or fleet and your work…
mastering-python-skill
Modern Python coaching covering language foundations through advanced production patterns. Use when asked to "write Python code", "explain Python concepts", "set up a Python project", "configure Poetry or PDM", "write pytest tests", "create a FastAPI endpoint", "run uvicorn server", "configure alembic migrations"…
kb-research
Structured approach to finding and synthesizing information from the user's knowledge base.
Agent Management
Create, edit, upgrade, delete, and audit agents.
gsd-check-todos
List pending todos and select one to work on.
gsd-manager
Interactive command center for managing multiple phases from one terminal.