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 fdueblab/Micro-Agent --skill financial_timeseriesgit clone --depth 1 https://github.com/fdueblab/Micro-AgentWrote 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/fdueblab/micro-agent/financial_timeseries)<a href="https://agentmods.dev/skills/fdueblab/micro-agent/financial_timeseries"><img src="https://agentmods.dev/badge/skills/fdueblab/micro-agent/financial_timeseries/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/fdueblab/micro-agent/financial_timeseries"><img src="https://agentmods.dev/badge/skills/fdueblab/micro-agent/financial_timeseries.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 | $0.00000 | $0.00633 |
| Opus 5.5 | $0.00000 | $0.00253 |
| Sonnet 5.5 | $0.00000 | $0.00127 |
| Haiku 4.5 | $0.00000 | $0.00063 |
Grade A, and why
financial_timeseries 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 22d 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.
What it actually says
你正在生成一个金融时间序列预测模型（汇率/价格/交易量/额度预测等）。时序任务有两条铁律，违反即失效：
铁律 1：防时序泄漏
- 切分：按时间顺序前 80% 训练 / 后 20% 测试，禁止
train_test_split(shuffle=True) - 特征：只用预测时点已知的信息——滞后特征（
shift(k)且 k ≥ 预测步长）、过去窗口的滚动统计、日历特征（星期/月末）；禁止当期特征、禁止全序列统计量（含未来数据） - 标准化：均值/方差仅用训练段计算
- 重采样禁止：SMOTE 会打乱时间结构，时序任务禁用；不平衡时用 class_weight
自检方法：对每个特征问"预测当天早上我能拿到这个数吗？"——答案是否则该特征泄漏。
铁律 2：必须对比朴素基线
朴素基线（下一期 = 最后一期值）在金融时序上极强，模型打不过它等于没有价值：
- 实现时同时计算：模型 MAPE vs 朴素基线 MAPE，两者并列写进 test_results 与 model_summary
- 模型接近随机游走数据（如日频汇率）时略逊于朴素基线是领域共识，如实汇报并说明，禁止掩饰
- 方向准确率 >55% 才有信息量（50% = 抛硬币），一并汇报
模型选型（按复杂度递增，单文件交付优先轻量）
- 朴素/季节性朴素（也是 no_training 约束下的合理方案）
- 线性回归 + 滞后特征（可解释，常胜过复杂模型）
- ARIMA（statsmodels，依赖声明带版本）
- GradientBoosting + 滞后/滚动特征（多变量主力）
禁止为单文件交付引入 LSTM/torch，除非用户明确要求。
输入防御（时序特有）
main_process 必须校验历史序列长度 ≥ 最小窗口（特征所需最大滞后阶数），不足时降级返回 {"prediction": None, "confidence": 0, "reason": "历史序列不足 N 期"}——禁止崩溃，禁止强行预测。
评估
MAPE 为主指标，RMSE/MAE 注明单位与量纲，方向准确率辅助。滚动验证（walk-forward）优于单次切分，条件允许时优先使用。
What ships with it
1 file 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.
- 22d ago First seen · 36 lines · 0 tokens per session scan A 1b3a438689b1
financial_timeseries is a skill published in the GitHub repository fdueblab/Micro-Agent (101 stars, last pushed 17d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 633 tokens. 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 2026-09-25.
Other skills, from other repositories
polymarket
Query Polymarket prediction markets. Check odds, find trending markets, search events, track price movements.
rag-knowledge
Work with the RAG knowledge base — ingest documents, run semantic search, manage collections, or add a sync source/connector (Google Drive, S3). Use when populating or debugging the knowledge base, tuning retrieval, or adding a new document source. This project uses {{ cookiecutter.vectorstore }} + {{…
billing-stripe
Work with Stripe billing — subscriptions, plans/prices, the Customer Portal, credits, usage metering, invoices, and webhook events. Use when changing plans, handling a new Stripe webhook, debugging a payment/subscription flow, or touching credit balances and usage.
rag-query-handling
Best practices for answering financial questions using RAG over bank statement data.
auto-trader-design
Use this skill to generate well-branded interfaces and assets for Auto Trader (multi-market AI-assisted trading workspace for KR / US / crypto), either for production work or throwaway prototypes, mocks, and screenshots. Contains essential design guidelines, color + type tokens, component specs, page patterns, and a…
vandalizer-api
Talk to the Vandalizer Management API — /api/mgmt/v1. Use when the user says "/vandalizer-api", asks to query the management API, pastes a vklive… key, asks to check Vandalizer dev/prod stats, validation runs, workflows, audit log, or wants to run a workflow/validation/extraction via the management surface.