Command·Engineering kit·plugin · v1.3.0·EC35
/database
Describe a PostgreSQL problem in plain words and it picks one database prompt from this library, says why, and runs it read-only when it can reach the code and database. In Codex it's $database.
Command
/database
The engineering-kit plugin in AtlasScope/agent-kit: the /engineer and /database routers plus 14 of Matt Pocock's skills, each read in full, pinned to a reviewed commit, and set to run only when you type them. Install it in Claude Code or Codex, or enable it for cloud sessions in a repo's .claude/settings.json.
Skills it uses
Related prompts
- Make this database easier to trust and operatePromptThe broad starting point when the database feels hard to trust or costly, and you can't yet name one problem.
- Can we independently prove these numbers?PromptImportant figures may be wrong even though their lineage is documented.
- Find data that is internally consistent but wrongPromptEvery check passes, but totals or outputs still look implausible.
- Find where records silently change identityPromptSources use aliases, changing identifiers, share classes or inconsistent names for the same thing.
- Reproduce what the database believed in the pastPromptPast reports, backfills, restatements or changed classification rules matter.
- Map every place that writes to the databasePromptSeveral jobs, functions, APIs or agents change the same data, or nobody is sure who owns it.
- Stop rewriting data that hasn't changedPromptUpserts and updates run at high volume, imports are costly, or vacuum and write-ahead log churn look high.
- Is our ingestion process more complicated than necessary?PromptThe import pipeline is long, slow to correct, or does the same work at several stages.
- Can we rerun everything without damaging the data?PromptImports get retried, schedulers overlap, files are replayed, or runs fail halfway.
- Find where write logic should livePromptBusiness rules are split across SQL, triggers, edge functions, server routes and workers.
- Which database objects no longer earn their keep?PromptThe list of views, functions and triggers has grown hard to reason about.
- Are we storing the same truth multiple times?PromptCaches, aggregates, materialized views or derived columns drift from the raw facts.
- Can the schema be easier to understand without losing depth?PromptNew developers or agents keep misunderstanding names, concepts or relationships.
- Find the true sources of database wastePromptThe bill is high, CPU or disk usage is high, storage keeps growing, or you suspect bloat.
- What will become expensive as the database grows?PromptFact or history tables are growing fast, or you need to plan capacity.
- Can we rebuild this database from its sources?PromptYou worry about disaster recovery, production-only state, or whether the data could be rebuilt.
- Are our database changes safe before we merge?PromptA migration, or a change to functions, views, RLS, types or column meaning, is about to ship.
- Which decisions should an expert agent make without me?PromptAgents keep asking you to classify fields, match records or pick between sources.
More in Engineering kit.