Skip to content

Tibsfox/gsd-skill-creator / examples/skills/gsd-meta/portable-schema-generator

portable-schema-generator

Verified

Emit Postgres.sql and SQLite.sql from a single schema spec so tools work across both drivers without duplicating DDL by hand. Use when designing a schema that needs to support both shared Postgres deployments and zero-config SQLite. Reduces two-file sync burden to a single source edit.

Portable Schema Generator

Two-driver schemas mean two migration files. Hand-keeping them in sync drifts fast. This skill captures the translation rules and the minimal tooling that keeps them aligned.

When to Use

  • Your tool needs to work on both Postgres and SQLite
  • You want fresh-project onboarding to be zero-config (SQLite default) while shared-infra deployments keep Postgres
  • A single schema change should land in both driver files with no manual diff-copying

Core Differences

ConcernPostgresSQLite
SchemasCREATE SCHEMA foo; SET search_path TO fooNamespace-flat — prefix names instead
AutoincrementBIGSERIAL PRIMARY KEYINTEGER PRIMARY KEY AUTOINCREMENT
TimestampsTIMESTAMPTZ DEFAULT now()TEXT DEFAULT (strftime('%Y-%m-%dT%H:%M:%fZ','now'))
BooleansBOOLEANINTEGER (0/1)
Enums via CHECKCHECK (col IN ('a','b'))Same — both support inline CHECK
Foreign keysOn by defaultPRAGMA foreign_keys = ON required at connect
Updated-at triggerCREATE TRIGGER ... EXECUTE FUNCTION touch_updated_at()CREATE TRIGGER ... BEGIN UPDATE ... END
Cascading dropsON DELETE CASCADESame, but only enforced with foreign_keys = ON
Reserved-word casingCase-insensitiveCase-insensitive, but be consistent

Translation Rules

When authoring the Postgres version first:

  1. Strip schema prefix for SQLite — CREATE SCHEMA IF NOT EXISTS foo; and SET search_path are removed. Keep bare table names.
  2. Replace BIGSERIAL with INTEGER PRIMARY KEY AUTOINCREMENT.
  3. Replace TIMESTAMPTZ with TEXT; default becomes (strftime('%Y-%m-%dT%H:%M:%fZ','now')).
  4. Replace BOOLEAN with INTEGER (values 0/1 at insert time — use an adapter coercion).
  5. Drop RESTART IDENTITY on truncates (SQLite uses DELETE FROM sqlite_sequence WHERE name = 'table' instead).
  6. Rewrite function-style triggers to inline BEGIN ... END trigger bodies.

File Layout

migrations/<feature>/
  001-init.postgres.sql
  001-init.sqlite.sql
  002-next.postgres.sql
  002-next.sqlite.sql

Numbered pairs. The adapter picks the right one based on cfg.db.driver.

Adapter Runtime Responsibilities

The query adapter at runtime should:

  • Rewrite $N placeholders → ? for SQLite
  • Strip release_history. schema prefix for SQLite
  • Translate now() → strftime(...), ::type casts → drop
  • Coerce JS booleans → 0/1 on param bindings
  • Provide a portable truncate(table) helper

Reference implementation: tools/release-history/db.mjs in this repo.

Workflow

  1. Author the Postgres version first (richer type system reduces ambiguity).
  2. Translate row-by-row using the rules above into the SQLite version.
  3. Test both: apply each migration to a disposable DB, run schema-check script against the adapter.
  4. Commit both files together; never commit one without the other.

Anti-patterns to avoid

  • Single mixed-syntax file — both drivers will try to parse it and one or both will fail.
  • Runtime translation of whole migration files — too many edge cases.
  • Letting the Postgres file diverge "just for now" — alignment rots quickly.

Example (from this repo)

See migrations/release-history/001-init.postgres.sql and migrations/release-history/001-init.sqlite.sql. Six tables (release, feature, metric, retrospective, lesson, publish_target). Same semantics, driver-idiomatic DDL.

Related

  • env-setup — database credentials and .env conventions
  • file-operation-patterns — migration file naming discipline
writing-skillsobra/superpowersUse when creating new skills, editing existing skills, or verifying skills work before deploymentStars: 322kUpdated Claude CodeCodexVerifiedInstall
xlsxanthropics/skillsUse this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .xltx, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like "the xlsx in my downloads") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.Stars: 197kUpdated Claude CodeCodexVerifiedInstall
shipping-and-launchaddyosmani/agent-skillsPrepares production launches. Use when preparing to deploy to production, or when asking what needs to be in place before shipping. Use when you need a pre-launch checklist, when setting up monitoring, when planning a staged rollout, or when you need a rollback strategy.Stars: 119kUpdated Claude CodeCodexVerifiedInstall
childhoodthedotmack/claude-memBabysit a shipped release after merge, publication, or deployment. Verify published packages, production behavior, installed plugins, and late CI or user reports until the observation window is clean; use babysit for PR readiness before release.Stars: 118kUpdated Claude CodeCodexVerifiedInstall