Home Learn Passport Live MVP Rust Funding Community Blog Papers GitHub ↗
Learn · Argorix Lang

Learn Argorix

From zero to a governed, verifiable agent — using only a stable Rust toolchain. No SaaS, no keys, fully offline.

Start here

Four steps to a governed agent

Each step is deterministic and offline. Nothing reaches a provider until governance allows it.

01

Build the toolchain

Compile the compiler, the VM and the conformance runner from source.

cargo build --workspace
02

Write a contract

Declare a module, agents, a passport and a Policy v2 block in a .argx file.

vim runtime.argx
03

Compile & verify

Validate semantics and verify the emitted bytecode before running anything.

argorixc verify-bytecode runtime.argx
04

Dry-run & inspect

Run the VM in dry-run, then read the exported security report and evidence bundle.

argorix-vm run runtime.argbc.json --dry-run
Core concepts

The model in six parts

Every concept maps to a declarative block the compiler, runtime and policy system can check.

Identity

Agent Passport

A technical, legal and security ID: who the agent is, who provides it, its jurisdiction, risk and evidence.

Open Passport
Governance

Policy v2

require / deny / on-violation rules, evaluated before any action. Fail-closed by default.

Execution

Execution Modes

dry_run, simulated and the guarded sandboxed_external — the runtime only does what governance permits.

See Live MVP
Evidence

Evidence Bundle

A portable, offline-checkable bundle with SHA-256 digests over canonical serialization.

Boundary

Adapters & Providers

Providers are non-executable by default; secrets are referenced, never read.

Defense

Threat Model

Named threats (e.g. prompt-injection) mapped to guards that hold turns for review.

A first contract

What .argx looks like

A minimal governed agent — declarative, compiled to bytecode, checked before it runs.

runtime.argx
module ArgorixChatbotRuntime

agent AssistantAgent {
    receives UserPrompt
    capabilities { trace.write }
}

passport AssistantPassport {
    global_id   = "argx:agent:01HZX9CHATASSISTANT"
    risk_level  = high
    jurisdiction = "CL"
}

policy ChatbotPolicy {
    deny external.execute
    on-violation => review   // fail-closed
}

Keep going

The full language reference, examples and conformance suite live in the repository.