一键导入
sqlite
Embedded SQLite — sqlite3 CLI, pragmas, JSON1, WAL, migrations; pairs with cursor-mirror; server DBs use postgres.
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
菜单
Embedded SQLite — sqlite3 CLI, pragmas, JSON1, WAL, migrations; pairs with cursor-mirror; server DBs use postgres.
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Portable tokens of capability, identity, and access
The grammar rules that make MOOLLM's file system object-oriented. Plural directory names declare element type; UPPERCASE marker files declare interface exports (COM-style, minus the UUIDs); directories are implementation classes exporting every interface whose marker file sits at their root.
Mother skill for platform-descriptor sister skills. Defines what a BIOME is — a bounded region of an ecosystem (coexisting, exchanging, never isolated) for one platform you operate — and what files, subdirectories, and cross-biome bridges every daughter biome inherits.
A skill is documentation that learned to do things.
GNU Emacs as a stateful Lisp machine for agents — daemon, moo-* protocol, emacs.py router, emacs:// URLs, spoken grammar, play-learn-lift.
Schemapedia — schema plugins, families, gateways, formats.yml, mechanism_relations; self-object kernel; delegates to sibling skills.
| name | sqlite |
| description | Embedded SQLite — sqlite3 CLI, pragmas, JSON1, WAL, migrations; pairs with cursor-mirror; server DBs use postgres. |
| license | MIT |
| tier | 1 |
| allowed-tools | ["read","grep","glob","run_terminal_cmd"] |
| related | ["postgres","schema","cursor-mirror","datasette","robust-first"] |
| tags | ["moollm","sqlite","embedded","json1","wal","migrations","datasette"] |
Part of MOOLLM · skills/sqlite/
SQLite is an embedded SQL engine: library + file (or :memory:). No separate server process. MOOLLM uses the sqlite mechanism in schemapedia for the engine; the sql mechanism names the SQL language family shared with Postgres and others.
This skill is the wrapper around the SQLite ecosystem: the sqlite3 CLI, C API and language bindings, JSON1, WAL, migration runners, and file format documentation—not a fork of SQLite. For read-only web exploration and JSON API over .db files, use the datasette skill and schemapedia datasette mechanism (including the cursor-mirror export path).
Use postgres when you need server-side roles, connection pooling, Timescale, pgvector, or HA patterns.
sqlite3 ./app.db
Useful interactive commands:
.tables — list tables.schema [table] — show DDL.headers on / .mode column — readable output.dump — SQL text exportExit: CTRL+D or .quit
Metadata lives in sqlite_master (and internal schema tables). Inspect:
SELECT type, name, tbl_name, sql FROM sqlite_master ORDER BY type, name;
journal_mode — WAL is often right for read-heavy concurrent access; understand locking (writers still serialize).foreign_keys — PRAGMA foreign_keys = ON; for referential checks (must be enabled per connection in many drivers).synchronous — durability vs speed tradeoff; never weaken without understanding data loss risk.user_version — integer you own for migration versioning.Official pragma reference: https://www.sqlite.org/pragma.html
JSON1 adds SQL functions for JSON in columns — json_extract, json_each, etc.
https://www.sqlite.org/json1.html
SQLite uses dynamic typing with type affinity. STRICT tables (optional) reject invalid types on insert. Read: https://www.sqlite.org/stricttables.html
user_version or use a schema_migrations table — pick one convention per project.FTS5 is SQLite’s full-text engine for search indexes — separate virtual tables from ordinary rows.
https://www.sqlite.org/fts5.html
| Concern | SQLite | PostgreSQL |
|---|---|---|
| Concurrency | Single writer; WAL helps readers | MVCC; server tuning |
| Extensions | JSON1, FTS5 built-in | Timescale, pgvector, … |
| Deploy | Single file | Host + roles + backups |
Porting often requires type and locking semantics review — not a mechanical rename.
.db files)Datasette publishes SQLite files as a read-only web UI and JSON API using the datasette CLI, optional metadata YAML, and --crossdb for multiple databases. It does not replace sqlite3 for DDL or bulk writes. MOOLLM wires cursor-mirror to Datasette via export_datasette and reference/universal/datasette-metadata.yml — see the datasette skill for the full integration protocol.
Cursor stores SQLite databases locally. The cursor-mirror skill documents table shapes and YAML models — see schemas/mechanisms/cursor-mirror in the registry.
postgres — server Postgres operations and extensions.schema — schemapedia sqlite mechanism profile.cursor-mirror — introspection over Cursor SQLite data.This skill's directory (browse and fetch everything): skills/sqlite/