agentleFS
atsushi-greenatsushi-green/ds-ai-coding-skills/.claude/skills/analysis-reporting/SKILL.md

analysis-reporting

Skill86 starsChanged 51 days agoMITView on GitHub

Use this when summarizing analysis results, writing reports, documenting experiment outcomes, or presenting model evaluation — including structuring findings in Japanese with conclusions, facts, assumptions, interpretations, and caveats.

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

6 rules. About 0 are common, 5 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈5

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use this skill when summarizing analysis results, experiment outcomes, or model evaluation.

    line 8
  2. Very specific:

    When any statistical / ML / causal / simulation / optimization method was applied, add a 診断サマリー section between 事実 and 解釈, containing the …

    line 27
  3. Distinctive:

    Every report should include or reference:

    line 49
  4. Distinctive:

    Use business-friendly language for summaries intended for stakeholders.

    line 58
  5. Distinctive:

    Keep technical details available for reviewers (in appendix or linked notebook).

    line 59
  6. Distinctive:

    Avoid jargon when simpler terms suffice.

    line 60

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Skill: Analysis Reporting

Use this skill when summarizing analysis results, experiment outcomes, or model evaluation.

Language

  • Reports should be written in Japanese unless otherwise requested.

Structure

  1. 結論 — Start with the conclusion or key finding.
  2. 事実 — Present objective facts from the data.
  3. 仮定 — State assumptions made during analysis.
  4. 解釈 — Provide interpretations and implications.
  5. 制約・注意点 — Mention limitations, caveats, and possible bias.

A fill-in-the-blank report skeleton with all of the above, plus the reproducibility section, is in references/report-template.md.

Diagnostics Summary(診断サマリー)

When any statistical / ML / causal / simulation / optimization method was applied, add a 診断サマリー section between 事実 and 解釈, containing the table defined by the matching *-diagnostics skill:

診断項目実測値合格基準判定次アクション
  • 判定 is one of OK / 要対処 / 確認(人間の判断待ち). Every 要対処 needs a 次アクション.
  • The analysis code only saves the figures and prints (or saves) the diagnostic values as a table of metric names and numbers. Copy 実測値 from that output, then write 合格基準, 判定 and 次アクション in the report by reading the values and figures against the skill's criteria. Do not build verdict strings, criteria text or next actions with if-branches or f-strings in the analysis code (an assert that stops an invalid analysis, such as an SRM check, is fine).
  • Include the table even when every item is OK.
  • Link the figure directory (outputs/diagnostics/<YYYYMMDD-HHMM>_<skill>/) instead of embedding figures.

Required Context

Include the following when relevant:

  • データ期間: Date range of the analysis.
  • フィルタ条件: Filters applied to the data.
  • サンプルサイズ: Number of records or observations.
  • 指標定義: How key metrics are calculated.

Reproducibility Notes

Every report should include or reference:

  • 入力データパス: Path to input data used.
  • クエリ/スクリプトパス: Path to the query, notebook, or script that produced the results.
  • 出力パス: Path to output artifacts (figures, tables, reports).
  • 実行コマンド: Command used to generate results.

Tone

  • Use business-friendly language for summaries intended for stakeholders.
  • Keep technical details available for reviewers (in appendix or linked notebook).
  • Avoid jargon when simpler terms suffice.
  • Be precise about what the data shows vs. what is inferred.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.