Server Os

io.github.ANAMIZEDPublished by GitHub account ANAMIZEDv0.1.0Updated Oct 24, 2026

Fail-closed agentic OS MCP: agents as processes with cost, governance, observability, skills.

VerifiedSTDIODesktop onlyCloud & InfrastructureSecurity & MonitoringAI & ML

Overview

AI-generated overview

Runs an agentic OS locally, letting an assistant manage agents as processes with cost control, governance, observability and skills.

What you can do
Server OS presents agents as first-class processes inside a fail-closed agentic operating system. It exposes an MCP server entry point alongside a REST API, CLI, SDK, a web control plane and skill definitions, and it ships verification scripts covering API, cost, governance, multi-agent workflows and skills. Default operation uses an offline mock LLM, with a real model mode available through configuration.
When to use it
Worth considering when you want an assistant to orchestrate multiple agents with explicit budgets, policy enforcement and observability, or when you need a self-contained local runtime for experimenting with agent governance and multi-agent workflows.
Requirements
Runs as a local process over stdio. The README describes Docker Compose for the production runtime and a Python-based REST API, CLI, SDK and MCP entry point. Real model use requires setting SERVER_OS_LLM_MODE=openai and OPENAI_API_KEY; the default offline mock LLM needs no key. Local HTTP ports are used for the API and dashboard.
Before you install
The README includes donation addresses for USDC on Base, Ethereum and Solana and mentions paid support tiers, so treat any payment request as optional and unrelated to installation. Setting OPENAI_API_KEY sends prompts to an external model provider. The system runs agents that can act on your machine, so review its governance and least-privilege settings before use.

Installation

In SourceWeft

  1. Open Server Os in the dashboard and add it to a workspace.
  2. Enable the server for the chats that should use its tools.

Desktop only via STDIO. STDIO servers start a local process, so they need the SourceWeft desktop host.

Other MCP clients

Follow the launch instructions in the repository.

README

Server OS

[CI] [License] [Python 3.11+] [MCP] [SDK] [CLI] [API]

Autonomous Agentic Operating System — agents as first-class processes with cost control, governance, observability, MCP, SDK, CLI, and skills. Fail-closed. Verified end-to-end.

A senior engineer who has never seen this repository can, using only the source code and this README.md:

  1. Deploy the entire system (single command)
  2. Exercise every major feature
  3. Verify end-to-end correctness via automated checks

No prior context or tribal knowledge required.

Support Agentic OS Kernels ($99) · Agentic OS Cycle ($0.75) · Public Goods Support

Non-custodial USDC (preferred for agents)

NetworkAddressExplorer
Base0xD3d0E9eDAe3Ac7bb199a8EAA761BdA423b878438basescan
Ethereum0xD3d0E9eDAe3Ac7bb199a8EAA761BdA423b878438etherscan
SolanaETQwWf19axArsY493UfC6bxe2BmEzmzvCb58PPnC38Asolscan

Related: rui · LRSI · openmesha · OpenGOS · agenticarb · x402-cloudflare-starter

🚀 Web Control Plane (Hero Demo — Zero Install)

Canonical file: web/server-os.html (~82 KB)

Open locally:

bash
python -m http.server 8088 --directory web# http://127.0.0.1:8088/server-os.html

Or Chrome → File → Open File… → web/server-os.html.

Acceptance: open the file → OS tab → Golden evals → run → 12/12 passed.

Offline mock provider, policy, budgets, graph, terminal, and 12 golden evals. The HTML engine is DOM-free below the UI so the eval suite exercises the same path as interactive use.

Quick Start (Production Runtime)

bash
docker compose up --build# another terminalbash scripts/verify.sh

Default mode is offline mock LLM (deterministic, free). Set SERVER_OS_LLM_MODE=openai + OPENAI_API_KEY for real models.

Surfaces

SurfaceEntry
Web control planeweb/server-os.html (offline, self-verifying)
REST APIpython -c "from server_os.main import run; run()"
CLIserver-os status / server-os agents ...
MCP Serverserver-os-mcp
SDKfrom server_os.sdk import ServerOSClient
Multi-agent workflowsskills/multi-agent-workflow/
Skillsskills/*/SKILL.md
AGENTS.mdCoding-agent contract at repo root
CI.github/workflows/ci.yml

Verify contract

bash
bash scripts/verify.sh

Covers API, cost, governance, multi-agent workflows, SDK, CLI, MCP, skills, and AGENTS.md. 16 checks. All must pass.

Web parity check (manual, ~30s): open full web/server-os.html → run Golden evals → 12/12.

Design principles

  1. Least privilege by construction (capabilities from intent)
  2. Cost is a first-class resource
  3. Observable by default
  4. Fail closed
  5. Deployable with zero tribal knowledge

License

Apache-2.0

Source: README.md at commit f5039ae

Tools

0
SourceWeft has not started this package in its sandbox yet, so its tools are not known.

Version history

1
  1. v0.1.0LatestSep 16, 2026

Score

40% of the applicable checks met

How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.

  • Reached in the last 30 days25 pts

    Keep the endpoint answering the MCP handshake; SourceWeft probes it every 30 days.

  • Auth declared10 pts

    Declare the credential headers in the registry entry, or keep the endpoint answering the probe so its auth can be classified.

  • Tools listed15 pts

    Answer tools/list without a credential, or ship a package that starts in a sandbox without one.

  • DescriptionRequired15 pts
  • Logo10 pts
  • Documentation15 pts
  • License10 pts

    Declare a license in the repository or the registry entry.