Private preview for macOS

An AI data analyst coworker that lives on your laptop.

Reproducible answers from agents grounded in a catalog — on top of the data you already trust.

01 Interop

Agents that move with your data. Same expressions on your laptop and in your warehouse. Snowflake, BigQuery, Databricks, DuckDB.

02 Lineage

Cited by construction. Every answer carries its expression hash, the inputs it was derived from, and the chat that produced it.

03 Cache

Cached + hashed for speed. Same expression → same hash → instant. No re-execution, no re-billed tokens.

01 Local-first by design

On your laptop, with your keys.

A native Mac app that runs on your own LLM key and warehouse credentials. Bulk data stays put — only the aggregates an expression needs cross into the local sandbox.

xorq_snow 11 vars on this Mac

TASKS CATALOGS PROFILES

YOUR KEYS 11

ANTHROPIC_API_KEYsk-ant-••••••••••

SNOWFLAKE_ACCOUNT••••••••••••

SNOWFLAKE_PRIVATE_KEY••••••••••••••••

DATA RESIDENCY

Warehouse bulk rows stay here

aggregates →

Local sandbox compute on your Mac

no proxy · no token markup

02 Cited by construction

Unbreakable lineage.

Every answer ties back to a cataloged expression — its source, its content-addressed cache, its schema. Auditable today, reproducible in six months.

composed 51241c900965 flights

DATA LINEAGE

THIS ENTRY

51241c900965 · composed · 2 cols

COMPOSED FROM 1

SOURCE-FLIGHTS

CACHE

letsql_cache-snapshot-d22bf8d…0dfa7b · parquet/

SCHEMA 2

airportstring total_flightsint64

OPERATION GRAPH 2

engineread

engine RemoteTable 16 cols source: @9f425664259a

read Read 16 cols

03 An opinionated agent harness

A harness you can trust.

Claude Code and Cursor were built for code. Xorq routes every tool call through the catalog — every answer clears a data-quality and verification gate before you see it. The LLM is swappable; the harness is what we ship.

whats the heaviest penguin finished

ORCHESTRATOR

The heaviest penguin is a male Gentoo weighing 6,300 g — Biscoe island, recorded 2007.

DATA QUALITY QUALITY OK

SOURCE-PENGUINS n=344 · nulls body_mass_g=2 · range 2700–6300g valid · no impossible values

VERIFICATION VERIFIED

Confirmed verbatim from VERIFY_SOURCE-PENGUINS_HEAVIEST

orchestrator verifier quality-checker

04 Replay any answer

Spans, cache hits, replayable.

Open any answer for the spans, hits, and durations behind it. Replay a single tool call. Diff against last week.

replayed whats the busiest airport

RESULT TRACE

Spans15

Cache hits1

Hit rate100%

Span · waterfallDuration

catalog 959ms

to_pyarrow 684ms

_register_node 659ms

cache_hit 659ms

compile <1ms

05 The proof, in benchmarks

DABStep @dabstep

450 questions over payment data · accuracy %

Haiku baseline50%

Sonnet baseline75%

Haiku + xorq catalog84%

same model · same prompt · only the catalog changes

Smaller model. Better answers.

DABStep — 450 questions over payment data. Same model, same prompt. Only the catalog changes. Haiku + catalog beats Sonnet baseline in half the turns.

Read the full story →

06 The open-source core · Apache-2.0

An executable memory layer for data work.

Agents do real work, then throw it away — scratch scripts gone by the next run. Xorq is the open-source layer that keeps it: durable, content-addressed artifacts anyone can reproduce and reuse.

Star 510

Declarative Ibis expressions that run the same on your laptop or your warehouse. The catalog that powers Xorq Desktop.

XORQ CONSOLE / INBOX

How can I help with Inbox?

HAIKU

2 CATALOGS 16 TOOLS 9 SKILLS 3 QA GATES

CACHED RESULTS 17 ENTRIES / 341 KB

[LETSQL_CACHE]d86db314 1 KB11H AGO

[SNAPSHOT]6544f7b6 12 KB12H AGO

[SNAPSHOT]af2021a3 4 KB12H AGO

[SNAPSHOT]76e92a4b 4 KB12H AGO

[SNAPSHOT]6ff65fb9 6 KB12H AGO

07 Things people ask first

What is Xorq Desktop?

A native Mac binary. Point it at your data, pick a workflow, and an agent works against a shared catalog of cited expressions.

When can I use it?

Private preview. Waitlist for the desktop; the library (uv add xorq) ships today.

Do I need to migrate my data?

No. Connects to Snowflake, Databricks, DuckDB, S3, Postgres — whatever you already run.

What is the “catalog”?

An immutable store of executable expressions. Each entry: expression + metadata + cached result, addressed by hash. Agents compose new entries on top of old ones.

How is this different from a notebook + an LLM?

Notebooks accumulate output. Xorq accumulates work. Last week’s expression is still here, still addressable; the next agent picks up where the last one left off.

Two people, same question — conflicts?

Same expression → same hash → automatic dedup. Identity is content, not file path.

Linux / Windows?

Desktop is Mac-first. The library and Textual TUI run everywhere Python does.

08 Two paths to the same harness

Pick how you adopt.

Run our integrated harness, or layer xorq’s primitives into the harness you already use. Same catalog. Same open-source library.

01 Our harness private preview

Xorq Desktop

The integrated harness. Open the app, point it at your warehouse, ask. Agent, catalog, and lineage in one native binary. No terminal.

$ open Xorq.app

02 BYO harness open source

Claude Code × xorq

Already in Claude Code? Layer the same catalog, lineage, and guardrails in as a plugin. No app, no migration.

> /plugin marketplace add xorq-labs/claude-plugins
> /plugin install xorq@xorq-plugins

status

Agents come and go. What compounds is the catalog — composable memories the next agent builds on.

Not the agents. The work.

# the catalog is open source · the coworker is in private preview