NOVA AgentOS is a local-first, evidence-first, user-controlled desktop agent operating system that translates goals into executable tasks, invokes models and tools within clear workspace, permission, and budget boundaries, and validates completion using real files, test results, and traceable evidence.

Stars

6

7-day growth

No data

Forks

0

Open issues

0

License

No data

Last updated

2026-07-30

AI repository intelligence
FR-AI / ANALYSIS

Why it is worth attention

It addresses critical gaps in traditional AI assistants by ensuring actual file modifications, build/test execution, multi-agent governance, crash recovery, and evidence provenance, all unified into a single execution chain with a Proof-of-Done system.

Who it is for

  • Developers building AI-powered workflows
  • Researchers needing reproducible, verifiable agent outcomes
  • DevOps engineers managing automated task pipelines
  • Project managers requiring transparent task budgets and permissions

Use cases

  • Automate code review and test execution with independent verification
  • Orchestrate multi-agent research and parallel worktree tournaments
  • Enforce budget and permission policies on cloud or local agent tasks
  • Persist long-running missions with safe recovery and audit trails

Strengths

  • Evidence-first paradigm with Proof-of-Done, evidence ledger, and status truth
  • Granular security boundaries: workspace restrictions, per-round permissions, budget caps, and agent isolation
  • Support for multiple model providers (OpenAI, DeepSeek, Kimi, Ollama) and MCP stdio/HTTP extensions
  • Cross-platform availability: Windows (Electron & WPF), macOS Preview, and .NET 8 bridge

Considerations

  • Still in preview (v0.9.0-preview.29); not yet 1.0 GA, installer unsigned, auto-update off by default
  • macOS Preview is unsigned, not notarized, and feature-incomplete compared to Windows
  • No open-source license declared; code defaults to all rights reserved

README quick start

快速开始

Related repositories

Similar projects matched by category, topics, and programming language.

MoonshotAI
Featured
MoonshotAI GitHub avatar

Kimi-K3

Kimi K3 is an open-weight, 2.8T-parameter native multimodal agentic model with a 1M-token context window, designed for frontier coding, knowledge work, and reasoning tasks.

AI & Machine LearningAI Agents
3,348
lopopolo
Featured
lopopolo GitHub avatar

harness-engineering

Harness Engineering is a methodology for improving coding agent outputs by carefully crafting the environment around them—providing curated context, tools, and executable constraints that encode an organization’s nonfunctional requirements and cumulative lessons.

AI & Machine LearningAI Agents
2,390
mshumer
Featured
mshumer GitHub avatar

Claude-of-Duty

A browser-based first-person shooter built entirely with procedural generation and orchestrated AI agents, featuring 55k lines of Three.js/WebGL2 code and no art assets.

AI & Machine LearningAI Agents
1,234