Agentlas OS

<p align="center"> <a href="https://agentlas.cloud/studio"> <img src="assets/readme/agentlas-one-launch-film-v2.gif" alt="Agentlas One launch film" width="960"> </a> </p> <p align="center"> <a href="https://agentlas.cloud"> <img src="assets/agentlas-agent-lab-banner.svg" alt="Agentlas Agent Lab banner"> </a> </p> <h1 align="center">Agentlas OS</h1> <p align="center"> <strong>Build it or borrow it. The agents you create stay yours.</strong><br> Turn a plain-language request into a runnable agent or team, borrow specialists from the public Agentlas Hub, and keep the agents you create available through your private, owner-scoped Agent Cloud.<br> Run them through supported hosts for the LLMs you already use. Hephaestus is the open-source engine underneath. </p> <p align="center"> <strong>We are Agent Trust. Your agent is not a program. It is an asset. — Agentlas —</strong> </p> <p align="center"> _{An agent you create is not tied to one model workspace or computer. To retrieve it from Cloud elsewhere, install Agentlas OS on a supported host and sign in.} </p> <p align="center"> <a href="https://github.com/agentlas-ai/Agentlas-OS/releases/latest"> <img alt="Latest release" src="https://img.shields.io/github/v/release/agentlas-ai/Agentlas-OS?label=release"> </a> <a href="LICENSE"> <img alt="License: Apache-2.0" src="https://img.shields.io/badge/license-Apache--2.0-green"> </a> <img alt="LLMs" src="https://img.shields.io/badge/LLMs-Claude%20Code%20%7C%20Codex%20%7C%20Gemini%20%7C%20Antigravity%20%7C%20Cursor%20%7C%20DeepSeek%20%7C%20GLM%20%7C%20Ollama-black"> </p> <p align="center"> <a href="README.md">English</a> · <a href="README.ko.md">Korean</a> · <a href="README.zh-CN.md">中文</a> · <a href="README.ja.md">日本語</a> · <a href="README.hi.md">हिन्दी</a> </p>

Paste to Install

Paste this into the LLM you are using now, such as Claude Code, Codex, Gemini CLI, Antigravity, or Cursor:

Install Agentlas OS / Hephaestus from this GitHub repo:
https://github.com/agentlas-ai/Agentlas-OS

Run the repository's canonical one-touch installer so the durable host command
adapters are installed too; do not stop after only installing the marketplace
plugin:
curl -fsSL https://raw.githubusercontent.com/agentlas-ai/Agentlas-OS/main/scripts/install-all-runtimes.sh | HEPHAESTUS_INSTALL_GLOBAL_ROUTER=1 bash

For Claude Code, confirm that `hephaestus@agentlas-core-engine` is enabled at
user scope and that `~/.claude/commands/agentlas.md` exists. Do not report the
install complete unless a bare `/agentlas` command will be available in the
next session. At the end, confirm the active plugin, command surface, and global
routing status, then tell me to restart the host or run `/reload-plugins`.

Use this when you are already inside an LLM and want the Agentlas command surface active there. For direct shell commands, see the install methods below.

<p align="center"> <a href="https://agentlas.cloud/desktop"> <img src="assets/readme/agentlas-desktop-hero.png" alt="Agentlas Desktop dashboard with local agents, owner-private Agent Cloud, Hub specialists, connected model hosts, and automations" width="960"> </a> </p> <p align="center"> _{Build, own, borrow, and run agents across your local workspace, private Agent Cloud, and the public Agentlas Hub.} </p>

Agentlas Desktop in motion

<table> <tr> <td width="42%" valign="middle">

Build an owned agent

Describe the work in plain language. Agentlas classifies the request, runs the interview and research gate, generates the package, verifies it, then asks whether to keep it only on this computer or save it privately in Agent Cloud for restore on another signed-in Desktop.

Agentlas Desktop →

</td> <td width="58%"> <a href="https://agentlas.cloud/desktop"><picture><source srcset="assets/readme/feature-wall/workflow-build-pipeline.gif" type="image/gif"><img src="assets/readme/feature-wall/workflow-build-pipeline.jpg" alt="Building an owned agent package from a plain-language request in Agentlas Desktop" width="100%"></picture></a> </td> </tr> <tr> <td width="42%" valign="middle">

Orchestrate a team

Combine local agents and borrowed Hub specialists into one orchestrator. Roles stay explicit while Agentlas manages routing, handoffs, and review boundaries.

Explore Agentlas OS →

</td> <td width="58%"> <a href="https://agentlas.cloud/models/hephaestus"><picture><source srcset="assets/readme/feature-wall/workflow-make-group.gif" type="image/gif"><img src="assets/readme/feature-wall/workflow-make-group.jpg" alt="Composing local and Hub agents into one orchestrated team" width="100%"></picture></a> </td> </tr> <tr> <td width="42%" valign="middle">

Run and verify locally

Use the model account or API key you choose. Your current host performs the work under its local files, tools, credentials, permissions, and verification rules.

Read the trust model →

</td> <td width="58%"> <a href="https://agentlas.cloud/docs/trust/agent-trust"><picture><source srcset="assets/readme/feature-wall/workflow-run.gif" type="image/gif"><img src="assets/readme/feature-wall/workflow-run.jpg" alt="Running and verifying an Agentlas agent under the local host permission boundary" width="100%"></picture></a> </td> </tr> </table> <p align="center"> <a href="#build-borrow-own">Build · Borrow · Own</a> · <a href="#why-agentlas-os">Why Agentlas OS</a> · <a href="#paste-to-install">Paste to Install</a> · <a href="#agentlas-desktop-in-motion">Desktop Demo</a> · <a href="#why-not-just-make-a-claude-agent">Why Not Just A Claude Agent?</a> · <a href="#all-install-methods">All Install Methods</a> · <a href="#the-command-surface">Command Surface</a> · <a href="#new-in-v110--the-briefing-interview-engine">New in v1.1.0</a> · <a href="#the-os-subsystems">Subsystems</a> · <a href="#where-this-fits">Product Surfaces</a> · <a href="#built-for-owned-agent-operations">Owned Agent Operations</a> · <a href="#what-it-builds">System Packaging</a> · <a href="#docs-by-goal">Docs Registry</a> </p>

---

Build, Borrow, Own

An agent you create should remain an asset you can move, rather than a setting trapped in one chat, one model-vendor workspace, or one computer. Agentlas separates three jobs that ordinary agent builders blur together:

This is the public Agent Trust contract: a portable, owner-scoped, inspectable, and restorable package contract—not a claim of regulated financial or legal trust services.

| Value | What Agentlas does | Entry point in an external LLM host | | --- | --- | --- | | Build | Compiles a plain-language request into a runnable single-agent or team package with roles, tools, memory boundaries, permissions, routing, and verification contracts. | /agentlas build | | Borrow | Finds public Hub specialists and brings the selected runtime bundle into your current Agentlas host. The publisher's private source work is not copied into your workspace. | /agentlas hub (Hub only) or /agentlas network (Local + Cloud + Hub) | | Own | Keeps agents you create in a private, owner-scoped Agent Cloud so you can retrieve and call them again after changing models or computers. | Choose private Agent Cloud at /agentlas upload, then retrieve with /agentlas cloud |

Portable package, local execution

Describe the work
  -> build a portable agent or team
  -> save it to my owner-scoped Agent Cloud
  -> install Agentlas OS and sign in on another supported host
  -> retrieve it with /agentlas cloud
  -> my chosen model and current host execute the work

Agent Cloud stores and retrieves the owner's package; it is not a hosted LLM that completes the work on the server. When you call a package, your selected model and current host runtime execute it under that host's permission and safety model. Credentials, local files, and machine-specific permissions do not travel with the package—you configure those separately on each computer.

The package contract — what every build must emit

An agent package is not a worker with capabilities. It is a method document. Three layers, and conflating them is the single most expensive mistake this project has made:

| layer | what it is | who supplies it | |---|---|---| | LLM | the worker: reasoning, language, general knowledge | you (BYOM) | | Runtime | the facilities: read the web, write files, run a shell, drive a browser | the host machine | | Agent | the work manual: procedure, judgement rules, source priority, input/output contracts, stop conditions | the package |

"Can this agent read the web?" is a category error. The runtime reads the web. The only question a package can answer is what method does it carry, and what must the machine be able to do for that method to run.

package-contract.json is the machine-readable list of artifacts every build emits, and scripts/verify-generated-package.sh <folder> is what enforces it. A build that omits a required artifact fails; it does not ship. Four of those artifacts carry the routing contract:

contracts/intake.schema.json     what a requester must hand over before work starts
contracts/output.schema.json     what the requester ends up holding
contracts/output.example.json    one real instance, validated by a JSON Schema
                                 validator at publish time — never by a model
.agentlas/brief.json             the compiled resume, schemaVersion agentlas.brief/1

Direction lives in the filename because nothing else ever marked it, and a schema whose direction must be guessed cannot be matched against a request.

Two rules bind every enum a build writes. Both were paid for in production:

1. **Any enum reachable from matching