OMGDB

LOCAL-FIRST · AI-NATIVE · v0.1 EARLY ACCESS

OMGDB

The local-first database agents can actually understand.

Store documents, Markdown, vectors, schemas, history, and debug metadata in one inspectable database. Built for AI-native apps, coding agents, and offline-first tools.

★ STAR ON GITHUB READ THE DOCS ⌘ TRY THE CLI →

~/projects/app — omg

// THE PROBLEM

Agents should not have to guess your database.

Most databases were designed for apps, admins, and query engines. AI agents need more: schema context, query explanations, safe mutations, semantic retrieval, and structured errors they can repair from.

?"Where is the schema?"

?"Why did this query return nothing?"

?"Which docs explain this feature?"

?"Can I change this safely?"

?"Are these embeddings stale?"

// WHAT OMGDB DOES

One local store. Five native layers.

01

Document Database

Flexible JSON-like documents, Mongo-style filters, aggregation, validation, and transactions.

02

Markdown Knowledge

Import docs, specs, notes, and AGENTS.md-style project context as structured, queryable documents.

03

Vector Search

Generate and search embeddings locally, with provenance and stale-state tracking.

04

Agent Debug APIs

describe, inspect, explain, validate, dry-run, rollback, and query debugging built into the database.

05

MCP Native

Expose database tools directly to coding agents through a first-class MCP surface.

// DEMO WORKFLOW

From empty folder to agent-ready database.

build.sh

$ omg init app.omgdb
$ omg create --from schema.yaml
$ omg import-md ./docs
$ omg find users '{"role":"admin"}'
$ omg explain users '{"age":{"$gte":18}}'
$ omg ask "which docs describe token expiry?"

SCHEMA SUMMARY

3 collections · users, docs, sessions · 12 indexes

MATCHING DOCUMENTS

4 docs matched {"role":"admin"}

CITATIONS

auth/tokens.md §3 · security.md §1

QUERY EXPLANATION

full scan → predicate age ≥ 18 → 812 rows

SUGGESTED INDEX

CREATE INDEX users.age — est. 9x faster

// AGENT-NATIVE

Built for tools that think before they write.

Self-Describing

Produces an agent-readable manual of its collections, schemas, indexes, vector state, and constraints.

Why-No-Results

Debug empty queries by showing which predicate failed and what nearby values exist.

Dry-Run Mutations

Preview changes before applying. Show matched documents, before/after samples, and risk notes.

Undoable Changes

Track agent writes with provenance and rollback metadata.

Context Packs

Return token-budgeted context for LLM tasks instead of dumping raw documents.

// WHY NOT GLUE CODE

Because agent infrastructure should not be glue code.

FEATURE

TRADITIONAL STACK

OMGDB

Local embedded DB

Usually no

Yes

Documents

Yes

Markdown-aware storage

Custom code

Native

Vector search

Separate service

Built in

Agent introspection

Custom prompts

Native

MCP tools

Custom server

Built in

Debuggable op-log

Rare

Canonical design

// DEVELOPER API

Two ways in. Both inspectable.

const db = await omg.open("app.omgdb")

await db.collection("docs").insert({
  title: "Auth Design",
  body:  markdown
})

await db.collection("docs").semanticSearch({
  query: "token expiry rules",
  k: 5
})

await db.explain({
  collection: "users",
  query: { age: { $gte: 18 } }
})

$ omg describe app.omgdb
# collections, schemas, indexes, vector state →

$ omg validate app.omgdb
# constraint + provenance check →

$ omg debug-query users '{"role":"admin","age":{"$gt":90}}'
# why-no-results: predicate age > 90 failed →

// TECHNICAL CREDIBILITY

Readable by default. Fast when it matters.

▸ canonical NDJSON operation log

▸ rebuildable binary cache

▸ deterministic dumps

▸ crash recovery

▸ local embeddings

▸ Markdown section indexing

▸ vector provenance

▸ MCP tool schemas

▸ Rust core, thin TypeScript layer

▶ INSERT COIN

Build local AI apps without teaching your agent the database from scratch.

★ STAR ON GITHUB INSTALL CLI READ ARCHITECTURE

OMGDB is early, local-first, and built in the open.