Tarski documentation
How to build an expert system on Tarski: write rules and scenarios, connect LLMs as sensors and deciders, and deploy to Tarski Cloud. These docs cover v0.5.9-preview.
Fast paths
Quickstart
Install the CLI, scaffold an app, and run verify on it.
What is Tarski?
What an expert system on Tarski is and how the pieces fit together.
The Tarski language
Relations, rules, invariants, and relay namespaces.
Client SDK
Append observations, run queries, and subscribe to live results from your app.
Start here
What is Tarski?
Tarski is a platform for building expert systems. This page explains what that means and how the pieces fit together.
Quickstart
Scaffold a Tarski app, run verification to check that its contract holds, and deploy the same app to the cloud in about five minutes.
Installation
Install the tarski CLI, pin versions, keep it updated, and understand what runs locally versus in the cloud.
Project structure
A Tarski app is a directory of text files interpreted by one binary, and this page explains what the manifest, ontology, mappers, helpers, queries, and fixtures are each responsible for.
Examples
Each Tarski example is a working app that teaches a specific containment or verification behavior, with fixtures that cover its unsafe paths as well as its happy paths.
Concepts
The observation-first pipeline
Observations are the only canonical evidence of what happened, and everything Tarski believes or does is computed from an observation history under one evaluator.
Facts, decisions & intents
How Tarski turns fallible evidence into trusted state and authorized action through candidates, proposals, the decision namespace, and the graph-cut safety boundary.
Worldviews & lineages
The identities that keep observation history, semantic meaning, package contents, and mapper output separate, and why effects bind to exactly one of them.
Provenance
Provenance is the explanation graph that shows why every candidate, decision, fact, intent, and effect exists, traced back to the observations that support it.
Determinism & replay
What Tarski's guarantees claim (byte-exact replay, dual-engine parity, query determinism, index neutrality, and hosted/local parity) and how each one is proven.
Build
The Tarski language
A strict Datalog subset for declaring ontologies, with relations, rules, decisions, invariants, and reserved namespaces that the loader checks to enforce containment.
Mappers & observations
The deterministic boundary where raw events become typed evidence, covering Rhai mappers, pure helpers, and deterministic effect providers.
Entities
One declaration expands into the whole record kit, with versioned latest-wins state, fail-closed conflicts, soft delete with revival, and generated read queries.
Write policies
Admit observations before append with explicit actor, payload, target, metadata, and idempotency bindings.
Scenarios & fixtures
Fixtures pair deterministic observation timelines with exact expected worlds and form the proof model behind tarski verify, blessing, query expectation sidecars, and proof artifacts.
Codegen & portable fixtures
Generate TypeScript contracts from an app package and move canonical observation worlds between clean evaluators.
Effects & reconciliation
How Tarski performs external actions through declared capabilities, admission identity, retry policies, the effect lifecycle, work queues, and explicit reconciliation.
Declared queries
The named, typed, parameterized read surface over your derived worldview, covering authoring, exposure, validation, and the query engine that runs it.
Reactive queries
Keep a declared query's result current without polling, using durable per-query subscriptions, resumable deltas, held streams, and explicit rehydrates.
Workviews, Collectors & modules
Workflow-shaped inspection and acquisition without a workflow engine, using read models, dispatch policy, scenario-driven collection with completion seals, and composable modules.
Scheduling
Standing cron, interval, and one-shot schedules whose exactly-once fires enter the lineage as ordinary observations, so the lineage stays the only source of truth.
Storage
Content-addressed blob storage with backend invariance, quotas, and GC, plus semantic tiering that keeps observation history at object-storage cost.
Plugins & packages
Installable semantic packs for external systems, with designated namespaces, load-checked surface tiers, typed HTTP receipts, and locked package management.
Agents
Authoring with agents
Tarski ships no coding agent of its own; it provides the contract surface, diagnostics, and machine-checkable feedback that let any capable coding agent write and check your rules.
Agents, interfaces & models
Declare agent roles, typed session starts, model resources, review routing, and headless model consultations.
Runtime containment
How Tarski keeps runtime LLMs in a bounded role, using relay namespaces, decision gates, invariants, and coordination through a shared derived model.
The agent query relay
How an LLM agent reads the worldview through a proposal, a decision, an admitted intent, a governed effect, and results that re-enter as gated evidence.
Platform
CLI reference
The tarski binary's command families: scaffold, dev, verify, query, studio, cloud, inspection, and maintenance.
Client SDK
Install and use @tarski/client for observation, query, subscription, schedule, blob, session, and platform-builder contracts.
Studio
The browser app for inspecting what Tarski derives and runs, including observations, facts, intents, effects, provenance, queries, and cloud state.
Cloud
The managed platform, covering cloud-first development, deploys with receipts, hosted parity and health, provider bindings, and multi-tenant read models.
Platform-builder control plane
Provision tenant lineages and issue, rotate, list, and revoke delegated runtime credentials from a service backend.
Identity & tenancy
How Tarski keeps real users, agent sessions, and tenant organizations separate with verified identity at ingress, authorization as derived facts, and fail-closed boundaries.
Versions & releases
Release channels, immutable artifacts, version pinning, and what "preview" means for the platform today.
Feature status
An availability matrix for current, preview, and planned Tarski platform surfaces.