本文へ移動
cccskills
無料GitHub で公開

schema-and-cleanup

Fix data that has gone wrong and change shape without breaking what reads it: duplicates, nulls, inconsistent document shapes, a field that means two things, a column that has to change. Use when a report cannot be built because the data is not in the right shape, when duplicates, nulls or inconsistent shapes are producing wrong numbers, when a field is needed or has to change, before a migration, or when onboarding data from a new source that does not match the standard shape.

インストール方法を見る

含まれるファイル(1)

  • SKILL.md2.7 KB

SKILL.md(原文)

インストールする前に、エージェントに与えられる指示の中身を確認できます。

Schema and cleanup

Goal

The data means what people think it means, the mess is fixed at its source rather than swept, and nothing reading it breaks.

Procedure

  1. Establish what the field is supposed to mean from whoever writes it, before touching anything. A badly named field usually means two things, and guessing which one is how a cleanup destroys the half you did not know about.
  2. Measure the mess before fixing it. How many rows, which audiences, since when. A cleanup with no baseline cannot be shown to have worked, and this is the number you will be asked for.
  3. Find where the bad data comes from. A cleanup that does not fix the source runs again next month, and then every month after that.
  4. Check who reads it before changing shape. memory/conventions/systems.md records what reads each table from outside the application, which is where the reporting scripts and integrations live. Those break silently.
  5. Make it reversible. A backup, a new column alongside rather than a change in place, a migration that can be run twice without doing damage.
  6. Fix forward and backward as separate steps. Stop the new bad data first, then clean the history, and say which of the two you did. Cleaning history while the source is still producing mess is work that undoes itself.
  7. Verify against the counts you took at the start and report the before and after.

Boundaries

  • Never delete or overwrite production data without an explicit go ahead and a backup.
  • Never change a column or field shape without checking what reads it from outside the application.
  • Never run a migration you cannot roll back.
  • Never guess what a badly named field means. Ask whoever writes it.
  • Never clean history while the source is still producing bad data, unless somebody decided that deliberately and knows it will recur.

What to record

Write what each ambiguous field actually means into memory/conventions/systems.md once it is settled, along with anything reading it from outside the application. Record each cleanup with its before and after counts and whether the source was fixed, since an unfixed source is a scheduled repeat of the same work.

レビュー

まだレビューはありません。使ってみた感想をお寄せください。

同じリポジトリのスキル

概要と使いどころ

Plan or review software architecture through a small expert panel. Use when the user explicitly asks for architecture planning, system design, technical strategy, an architecture assessment, a design review, or a system health review. Do not use for ordinary PR review, debugging, or implementation.

日本語の概要は準備中です。原文の説明を表示しています。

nanocoai/nanoclaw-templates182026年9月16日 更新

Thorough, consistently formatted competitor and product research. Produces a structured "About" doc, a "Recent News" log, and a row in a tracking spreadsheet. Use whenever the user researches a company or competitor, builds or updates a competitor doc, compares products, or logs a competitor to the tracker.

日本語の概要は準備中です。原文の説明を表示しています。

nanocoai/nanoclaw-templates182026年9月16日 更新

Household assistant for a whole family. Runs a morning brief, plans meals and builds grocery lists (hunting the best prices), looks ahead at the week's logistics, stays on top of the kids' school, watches prices on things the family wants, and helps book appointments and reservations. Use it for anything that keeps the family's day or week running. Trigger even on implicit asks - "what's on today", "what do the kids have this week", "what should we make for dinner", "add milk to the list", "when's the dentist", "did anything come from school", "remind me about the permission slip", "watch this jacket and tell me if it drops", "book us a table Friday".

日本語の概要は準備中です。原文の説明を表示しています。

nanocoai/nanoclaw-templates182026年9月16日 更新

Reporting workflow for a working journalist, using web research and Apify's X scraper (social signals). Onboards the journalist, monitors their beat and builds digests, evaluates story pitches, finds and vets sources, and prepares interviews. Use for any newsroom task, such as "what's new on my beat", "is this pitch any good", "who should I talk to about X", "prep me for this interview", or "turn these notes into a draft".

日本語の概要は準備中です。原文の説明を表示しています。

nanocoai/nanoclaw-templates182026年9月16日 更新

Check that the scheduled data work behind the reports actually ran and produced something sensible: the daily reshaping scripts, the API pulls, the tables the reports read from. Use each morning before anyone opens a report, when a report looks wrong and nobody knows whether it is the data or the query, after a schema or API change, when a new audience or source is added to the pipeline, or when a scheduled job failed or finished suspiciously fast.

日本語の概要は準備中です。原文の説明を表示しています。

nanocoai/nanoclaw-templates182026年9月16日 更新

Write the SQL or MongoDB query that answers a question and keeps answering it: the right grain, no fan out, parameterised window, checked against something known. Use when a number is needed that no report carries, when an existing query returns something that looks wrong, when a query is slow or expensive, or when somebody asks for a one off pull.

日本語の概要は準備中です。原文の説明を表示しています。

nanocoai/nanoclaw-templates182026年9月16日 更新

nanocoai のスキルをすべて見る

このスキルの問題を報告する