OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Agent Security & Sandboxing

Emilia Protocol

Emilia Protocol is authority-control infrastructure for autonomous work. Its EMILIA Gate checks whether an action is authorized under a finite, customer-owned mandate before a protected executor runs, and the open protocol supports independently checkable evidence.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Emilia Protocol — OpenViST project profile in Agent Security & Sandboxing.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use EMILIA when an autonomous agent can invoke a consequential tool and authorization must be checked at the executor boundary, especially when approval needs to apply to exact action details. Gate must mediate every path to the protected provider credential for the deployment to constrain those actions.

Who it is for

  • Teams building or operating agents that can call consequential tools.
  • System owners who need approval bound to exact action details and enforced where provider credentials are controlled.
  • Reviewers who need evidence and conformance materials for an authorization boundary.

Documented uses

  • Constrain an agent’s permitted work to a finite owner-defined mandate.
  • Require fresh approval for selected actions while allowing work within mandate limits.
  • Bind approval to exact action details and reject changed or replayed calls on covered paths.
  • Verify authorization evidence under trusted keys and rules.
  • Assess the documented authorization boundary using the repository’s example and conformance materials.
Before you choose

Requirements & limits

The public GitHub repository is identified as Apache-2.0 licensed. The documented local example requires Node.js 20.19 or newer and npm.

  • Gate enforces authority only on paths that the deployment completely mediates; a bypass path is outside that control.
  • The local payment example uses generated keys, in-memory consumption and a mock tool, so it is illustrative rather than a production payment flow.
  • Approval evidence does not prove that an action executed successfully or caused its intended effect; uncertain provider outcomes require reconciliation.
  • The project is authority infrastructure, not a replacement for identity providers, workload identity or policy engines.
  • The repository describes both CLI and server platform forms and identifies TypeScript as its primary language.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Run the local payment demonstration

Start the repository’s mock payment example to inspect mandate checks and exact-action binding.

Before you start

  • Node.js 20.19 or newer
  • npm
  • A local copy of the public repository

Steps

  1. Clone the repository and change into its directory.

  2. Install dependencies with npm ci.

  3. Start the example with FAST=1 node examples/mcp/payment-server.mjs.

Expected outcome

The local example demonstrates refusal without approval, admission of a matching payment call once, and refusal of changed details, replay and forged evidence.

Guide limitations

  • This is a local demonstration using generated signing keys, in-memory consumption and a mock payment tool.
  • It does not perform a real human approval ceremony or move money.
  • Production requires enrolled credentials, durable shared state and Gate mediation on every path to the protected provider credential.
  • A valid approval does not establish that bank details are legitimate.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
13 October 2026
Original website title
emiliaprotocol/emilia-protocol · Official homepage title · retained in its source language
How we review sourcesSuggest a correction