coding-pg

Owns PostgreSQL work: schema and DDL, migrations and lock levels, set-algebraic queries, functions and PL/pgSQL, indexes, RLS and tenant isolation, extensions (pgvector, PostGIS, TimescaleDB, pg_partman, pg_cron, pg_duckdb), logical replication and CDC, server tuning (vacuum, WAL, parallel query, JIT), EXPLAIN plans, wait-event and lock-contention diagnosis, and SQL embedded in TypeScript through @effect/sql-pg. Use when authoring, reviewing, migrating, or debugging any .sql file or SQL string and "lint my SQL". Provisioning and hosting a database instance belongs to pulumi.

Ir para a instalação

Repositório: bsamiee/Parametric_Forge
Última atividade na origem: 27 de julho de 2026 às 19:07
Idioma detectado do SKILL.md: inglês
Estrelas: 2
Forks: 0

Opções de instalação

Por padrão, está selecionado o prompt que primeiro revisa a origem. Você pode mudar para um comando direto ou baixar uma cópia local.

Instalar com Codex ou Claude Copie este prompt, cole no Codex, Claude ou outro assistente e deixe que ele revise a página da skill e instale para você.

Um comando direto ignora o prompt de revisão. Verifique a origem antes de executá-lo.

npx skills add https://github.com/bsamiee/Parametric_Forge --skill coding-pg

O comando permanece em uma só linha. Role horizontalmente para revisá-lo antes de copiar.

Prefere uma cópia local? Baixe os arquivos disponíveis atualmente no SkillsMP.

Revise os arquivos de origem

Leia o SKILL.md e os arquivos complementares exibidos pelo SkillsMP antes de decidir se vai instalar.

Explorador de arquivos

12 arquivos

SKILL.md

Instruções da origem · Visualização somente leitura

Voltar à pesquisa Voltar à instalação

[CODING_PG]

All SQL follows these governing principles:

[POLYMORPHIC] — one function/query per concern, generic over specific via parameter dispatch and dynamic SQL

[SET_ALGEBRAIC] — express operations as set transformations; zero row-at-a-time iteration

[STRONGLY_TYPED] — domain types, composite types, range types; zero untyped text columns for structured data

[PROGRAMMATIC] — variable-driven predicates, parameterized DDL, zero stringly-typed identifiers

[DECLARATIVE_FIRST] — constraints, generated columns, and RLS policies enforce invariants at the schema level; application logic is last resort

[SOURCE_CURRENT] — every example uses PostgreSQL semantics

[01]-[ROUTING]

[REFERENCES]: validation.md always loads; the rest load when the task matches.

[01]-VALIDATION: compliance checklist, Effect-SQL alignment, migration safety

[02]-DDL: schema design, domain/composite/range types, temporal constraints, generated columns, partitioning, lock levels

[03]-QUERIES: CTE algebra, MERGE, window functions, JSON_TABLE, recursive patterns

[04]-INDEXES: index type selection, partial indexes, covering indexes, maintenance

[05]-FUNCTIONS: polymorphic functions, custom aggregates, procedures, PL/pgSQL dispatch

[CODING_PG]

All SQL follows these governing principles:

[POLYMORPHIC] — one function/query per concern, generic over specific via parameter dispatch and dynamic SQL

[SET_ALGEBRAIC] — express operations as set transformations; zero row-at-a-time iteration

[STRONGLY_TYPED] — domain types, composite types, range types; zero untyped text columns for structured data

[PROGRAMMATIC] — variable-driven predicates, parameterized DDL, zero stringly-typed identifiers

[DECLARATIVE_FIRST] — constraints, generated columns, and RLS policies enforce invariants at the schema level; application logic is last resort

[SOURCE_CURRENT] — every example uses PostgreSQL semantics

[01]-[ROUTING]

[REFERENCES]: validation.md always loads; the rest load when the task matches.

[01]-VALIDATION: compliance checklist, Effect-SQL alignment, migration safety

[02]-DDL: schema design, domain/composite/range types, temporal constraints, generated columns, partitioning, lock levels

[03]-QUERIES: CTE algebra, MERGE, window functions, JSON_TABLE, recursive patterns

[04]-INDEXES: index type selection, partial indexes, covering indexes, maintenance

[05]-FUNCTIONS: polymorphic functions, custom aggregates, procedures, PL/pgSQL dispatch

[INDEX]

[LABEL]

[SYMPTOM]

[01]

STRING_TYPING

text for structured data (JSON, enum values, composite structures)

[02]

DUAL_COLUMN_RANGE

start_date/end_date instead of tstzrange with exclusion constraint

[03]

IMPERATIVE_BATCH

PL/pgSQL LOOP with row-at-a-time UPDATE instead of set-based MERGE/CTE

[04]

INDEX_SPRAWL

Redundant single-column indexes subsumed by existing composite indexes

[05]

NULLABLE_DEFAULT

Columns nullable without documented justification

[06]

TRIGGER_LOGIC

Business logic in triggers instead of MERGE RETURNING or generated columns

[07]

OFFSET_PAGINATION

LIMIT/OFFSET instead of keyset pagination for client-facing endpoints

[08]

SECURITY_DEFINER_LEAK

SECURITY DEFINER without SET search_path — search_path injection vector

[09]

STRINGLY_POLICY

RLS policy with hardcoded literals instead of current_setting()

[10]

FUNCTION_PROLIFERATION

Separate functions for each query variant instead of one polymorphic function

[11]

APPLICATION_SIDE_JSON

Fetching raw JSONB and parsing in application instead of jsonb_path_query

[12]

MANUAL_PARTITION

Hand-written partition creation instead of pg_partman lifecycle management

[13]

UNVALIDATED_CONSTRAINT

ADD CONSTRAINT ... NOT VALID without subsequent VALIDATE CONSTRAINT

[14]

IF_THEN_DISPATCH

PL/pgSQL IF p_op = 'get' THEN ... ELSIF instead of VALUES-based dynamic SQL

[15]

NONCOMPOSABLE_CAGG

PERCENTILE_CONT in hierarchical CAGG; non-composable across tiers

[16]

LEGACY_UUID

Non-ordered UUID generation on new ordered PKs where uuidv7() fits better

[17]

STALE_HEALTH_VIEW

Materialized views for real-time health monitoring instead of inline queries

[18]

EXCLUDE_OVER_WITHOUT_OVERLAPS

EXCLUDE instead of WITHOUT OVERLAPS PK/UNIQUE for temporal overlap

[19]

RAW_UUID_ID

Raw S.UUID for PK/FK instead of S.UUID.pipe(S.brand('EntityId'))

[20]

BARE_FOR_UPDATE

FOR UPDATE without SKIP LOCKED on batch/queue processing patterns

[21]

NULL_UNSAFE_ANTIJOIN

NOT IN (SELECT ...) instead of NOT EXISTS; NULL in subquery yields UNKNOWN

[22]

DISTINCT_OVER_EXISTS

SELECT DISTINCT on joined data; use EXISTS semi-join (avoids sort/dedup)

[INDEX]	[LAYER]	[MECHANISM]	[OWNS]
[01]	Identity	`uuidv7()`	PK generation — timestamp-ordered, B-tree friendly
[02]	Temporal	`WITHOUT OVERLAPS` + `PERIOD`	Range integrity — temporal PKs, FKs, exclusion constraints
[03]	Projection	Virtual generated columns	Computed fields — zero storage, instant `ALTER TABLE`
[04]	Write strategy	`MERGE ... RETURNING OLD/NEW`	Upsert + audit trail in single statement
[05]	Type safety	Domain types + composite types	Semantic column typing — branded scalars, structured records
[06]	Search	GIN + pg_trgm / pgvector	Full-text, trigram similarity, vector nearest-neighbor
[07]	Partitioning	Declarative + pg_partman	Time/hash/list partitioning with automatic lifecycle
[08]	Scheduling	pg_cron	In-database job scheduling — maintenance, materialized views
[09]	Security	RLS + SECURITY INVOKER	Row-level tenant isolation, least-privilege function execution
[10]	Observability	pg_stat_statements + auto_explain	Query fingerprinting, automatic slow-query plan capture

[INDEX]	[EXTENSION]	[OWNS]	[LOAD_WHEN]
[01]	`pgvector`	Vector storage, HNSW/IVFFlat/DiskANN indexes	Embedding search, similarity queries
[02]	`pgvectorscale`	DiskANN index, Statistical Binary Quantization	>1M vectors, memory-constrained
[03]	`pg_search`	BM25 full-text via Tantivy, `@@@` operator	Search-quality ranking, hybrid search
[04]	`pg_trgm`	Trigram similarity, GIN trigram indexes, `%` operator	Fuzzy text search, typo tolerance
[05]	`PostGIS`	Geometry/geography types, spatial indexes, ST_* funcs	Geospatial queries, proximity search
[06]	`TimescaleDB`	Hypertables, continuous aggregates, compression	Time-series ingestion, rollup queries
[07]	`pg_cron`	In-database scheduled jobs	Materialized view refresh, maintenance
[08]	`pg_partman`	Partition lifecycle management	Non-time-series partitioning
[09]	`pg_duckdb`	Embedded DuckDB, analytical acceleration, lake access	OLAP queries, Parquet/Iceberg reads
[10]	`pg_jsonschema`	JSONB validation via JSON Schema CHECK constraints	Structured JSONB columns
[11]	`btree_gist`	GiST equality operators for EXCLUDE constraints	Range + equality exclusion constraints
[12]	`bloom`	Bloom filter index for wide-table equality	>5 columns, arbitrary WHERE combos
[13]	`pg_stat_statements`	Query fingerprinting, execution statistics	Performance analysis, regression detect
[14]	`pgaudit`	Compliance-grade audit logging (session + object)	SOC 2, HIPAA, PCI-DSS compliance

coding-pg

Opções de instalação

Revise os arquivos de origem

[CODING_PG]

[01]-[ROUTING]

[CODING_PG]

[01]-[ROUTING]

[02]-[PARADIGM]

[03]-[CONVENTIONS]

[04]-[CONTRACTS]

[05]-[ANTI_PATTERNS]

[06]-[FIRST_CLASS_EXTENSIONS]

[07]-[VALIDATION_GATE]

coding-pg

Informações da origem

Opções de instalação

Revise os arquivos de origem

Mais deste repositório

Mais deste repositório

[CODING_PG]

[01]-[ROUTING]

[CODING_PG]

[01]-[ROUTING]

[02]-[PARADIGM]

[03]-[CONVENTIONS]

[04]-[CONTRACTS]

[05]-[ANTI_PATTERNS]

[06]-[FIRST_CLASS_EXTENSIONS]

[07]-[VALIDATION_GATE]