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 agentmods add skills/realnghon/data-scientist/analysis-workflownpx skills add realnghon/data-scientist --skill analysis-workflowgit clone --depth 1 https://github.com/realnghon/data-scientistWrote 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/realnghon/data-scientist/analysis-workflow)<a href="https://agentmods.dev/skills/realnghon/data-scientist/analysis-workflow"><img src="https://agentmods.dev/badge/skills/realnghon/data-scientist/analysis-workflow.svg" alt="Measured on agentmods" 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 | $0.00095 | $0.01687 |
| Opus 5 | $0.00048 | $0.00843 |
| Sonnet 5 | $0.00019 | $0.00337 |
| Haiku 4.5 | $0.00010 | $0.00169 |
Grade A, and why
data-scientist 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 3d 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 — 91 lines — stays where its author put it; the contents beside it link to each section on GitHub.
数据科学家
3 阶段快速分析流程。
流程
1. 数据摄入
- 读取数据:用 pandas 读 CSV/Excel/Parquet,不用 shell
- 环境:
python --version+ 测试 import pandas/numpy/scipy,能用就不重装 - 探查:每列先确认存在 + dtype;分类列再看
df[col].unique()。从 df 读,不靠记忆 - 大表:
os.path.getsize>100MB 时先nrows=5看 schema,再用usecols=只读所需列 - 多表 join:单表跳过。需要时记录 grain + join keys,用
ds_skill.shaping.audit_join预估匹配率,详见 data-shaping.md - 缺失预警:缺失率 >30% →
data_manifest标记high_missing=true,相关结论自动降一档置信度。永不 impute Y;impute X 须记录策略。详见 data-readiness.md - 产出:
data_manifest
data_manifest 最小格式(后续阶段只能引用这里记录过的列):
{"source": "path/to/data.csv", "n_rows": 10000, "target": "heart_attack",
"quality_score": "partial", "columns": [
{"name": "age", "dtype": "int64", "missing_rate": 0.0, "role": "feature"},
{"name": "smoker", "dtype": "object", "missing_rate": 0.0, "role": "group", "categories": ["smoker","non_smoker"]},
{"name": "alcohol", "dtype": "float64", "missing_rate": 0.393, "role": "feature", "high_missing": true}]}
字段规则:分类列必列 categories;missing_rate 0–1;缺失 >0.30 必带 high_missing;quality_score ∈ ok|partial|blocked。
2. 分析执行
- 选方法:查 method-registry.md
- 数据整形:按需 pivot/melt/aggregate,不单独成阶段
- 跑统计检验:任何"组间有无差异"的结论都要落到正式检验,不能只凭均值大小。
- 数值 Y by 组 →
ds_skill.analysis_methods.compare_numeric_by_group(自动选 Welch t / Mann-Whitney / ANOVA / Kruskal) - 分类 Y × 分类组 →
ds_skill.analysis_methods.compare_categorical(卡方,期望<5 自动转 Fisher,效应量 Cramér's V) - 连续关系 →
ds_skill.correlation.pairwise_correlation
- 数值 Y by 组 →
- 建模(按需):连续 Y →
ds_skill.regression,二分类 →ds_skill.classification,生存 →ds_skill.survival。详见 method-registry.md 第 5/6/8 章 - 混杂检查:报告前先问"有没有第三变量同时驱动两边"(如气温与病例数都随季节变化)。见 anti-patterns.md
- 画图:用下面的画图规则
- 产出:
evidence_matrix+ 图表
evidence_matrix 最小格式(每个发现一行):
[{"claim": "吸烟者高血压比例显著高于非吸烟者", "method": "chi_square_test",
"statistic": 64.2, "p_value": 0.0001, "effect": {"name": "cramers_v", "value": 0.40},
"n": 400, "tier": "reliable", "chart": "charts/smoker_hypertension.png", "caveats": []}]
字段规则:method 用 registry 名字;效应量必填;tier ∈ reliable|directional|unsupported,用到 high_missing 列或单一方法时不得为 reliable。
What ships with it
26 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.
- assets/report_template.md 4.1 KB
- references/anti-patterns.md 4.3 KB
- references/chart-catalog.md 3.0 KB
- references/data-readiness.md 3.2 KB
- references/data-shaping.md 4.2 KB
- references/manufacturing-playbook.md 4.7 KB
- references/method-registry.md 5.9 KB
- references/report-standard.md 5.9 KB
- scripts/ds_bootstrap.py 4.6 KB runs code
- scripts/ds_skill/__init__.py 2.5 KB runs code
- scripts/ds_skill/ab_validator.py 13 KB runs code
- scripts/ds_skill/analysis_methods.py 29 KB runs code
- scripts/ds_skill/anomaly.py 13 KB runs code
- scripts/ds_skill/bootstrap.py 6.3 KB runs code
- scripts/ds_skill/classification.py 16 KB runs code
- scripts/ds_skill/correlation.py 15 KB runs code
- scripts/ds_skill/plotting.py 22 KB runs code
- scripts/ds_skill/readiness.py 36 KB runs code
- scripts/ds_skill/regression.py 19 KB runs code
- scripts/ds_skill/shaping.py 17 KB runs code
- scripts/ds_skill/spc.py 31 KB runs code
- scripts/ds_skill/survival.py 24 KB runs code
- scripts/ds_skill/time_series.py 17 KB runs code
- scripts/profile_dataset.py 12 KB runs code
- scripts/run_full_workflow.py 18 KB runs code
- test-prompts.json 2.2 KB
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.
- 3d ago First seen · 91 lines · 95 tokens per session scan A c0c495a3ba32
data-scientist is a skill published in the GitHub repository realnghon/data-scientist (2 stars, last pushed 2mo ago), licensed MIT. It adds 95 tokens to every session and 1,687 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.
Other skills, from other repositories
agent-evaluation
Evaluate and improve Claude Code commands, skills, and agents. Use when testing prompt effectiveness, validating context engineering choices, or measuring improvement quality.
apply-anthropic-skill-best-practices
Comprehensive guide for skill development based on Anthropic's official best practices - use for complex skills requiring detailed structure.
analyse-problem
Comprehensive A3 one-page problem analysis with root cause and action plan.
create-skill
Guide for creating effective skills. This command should be used when users want to create a new skill (or update an existing skill) that extends Claude's capabilities with specialized knowledge, workflows, or tool integrations. Use when creating new skills, editing existing skills, or verifying skills work before…
create-rule
Use when found gap or repetative issue, that produced by you or implemenataion agent. Esentially use it each time when you say "You absolutly right, I should have done it differently." -> need create rule for this issue so it not appears again.
do-competitively
Execute tasks through competitive multi-agent generation, meta-judge evaluation specification, multi-judge evaluation, and evidence-based synthesis.