Annona 是一个开源的“主权执行内核”,用于决定、强制并记录 AI 智能体每一步可在哪里执行,默认本地运行、默认拒绝。

Stars

11

7 天增长

暂无数据

Fork 数

1

开放 Issue

0

开源协议

Apache-2.0

最近更新

2026-07-31

AI 仓库情报摘要
FR-AI / ANALYSIS

为什么值得关注

它将 AI 部署重新定义为按请求执行的策略而非单一架构选择,以运行时强制实现 fail-closed 放置,并通过哈希链账本让决策可验证。

适合谁使用

  • 处理敏感客户数据并需要符合 GDPR/EU AI Act 的组织
  • 在本地、欧盟云和前沿 API 之间部署 AI 智能体的团队
  • 需要可审计 AI 运行时策略的平台/安全工程师
  • 处理机密文件的法律与专业服务机构

典型使用场景

  • 当步骤接触受限文件时,在本地模型上运行智能体步骤,同时允许公开任务使用更便宜的前沿 API
  • 在没有允许的可用执行环境时,保留(hold)步骤而不是降级或改道
  • 导入 Claude Skills 并将其固定为仅本地运行
  • 在发送脱敏后的内容到前沿模型前,由本地 redactor 替换受限标识符

项目优势

  • fail-closed、默认拒绝的策略,配合单调污染追踪,不允许静默降级
  • 哈希链账本提供 verify/why/audit 命令,并有九项验收检查
  • 单一二进制、单一发布,可支持 detached、attached、appliance 三种拓扑而无需分叉
  • 分层架构通过构建期契约强制决策层无法直接调用 provider 适配器

使用前须知

  • 明确列出未完成项:无语法受限的工具调用、账本无外部锚点、排队步骤不会自动恢复
  • 旧配置路径在运行 annona policy init 之前仍为默认允许
  • 桌面安装包处于未签名 beta 阶段,需手动绕过操作系统警告

README 快速开始

Annona

The sovereign execution kernel for AI agents

Where it runs is a decision — and the decision is yours, enforced and recorded.

📐 Read the high-level design — components, the placement algorithm, the DGX appliance, the threat model, and the numbers it is judged by

🌐 akaion-ai.github.io/annona — the site · 🐳 docker compose up -d · 🐍 pip install annona

🐳 Container · 🖥️ CLI · 🧩 any OpenAI-compatible runtime — an-NO-na, the office that kept Rome fed


Rome imported its grain. Sicily, Africa, Egypt — the city could not feed itself, and everyone knew what that meant: whoever controlled the ships controlled Rome. So the Republic, and then the Empire, refused to leave it to the market. The cura annonae was a permanent office with a prefect at its head, and its job was to decide where the grain came from, which route it took, which granary held it, and who received it — and to keep the record of all four.

It was not built out of paranoia. It was built out of arithmetic: a republic cannot outsource what it cannot live without.

Your organisation is now in that position with compute. Every agent you deploy sends your material somewhere, and "somewhere" is currently decided by whichever provider a developer typed into a config file eighteen months ago.

Annona is the office. It is a daemon you install where your data already lives, and for every single step of every agent run it decides where that step is allowed to execute, enforces the decision, and writes it down.

$ annona why step_7f3a
step_7f3a  inference  HELD
  class        restricted  (working set touched /mnt/pratiche/2026/BG-114.pdf)
  rule         rules[0]  restricted → [local-gpu], on_unavailable: hold
  candidates   local-gpu (unhealthy: connection refused since 14:02:11)
  not chosen   frontier — max_class public  **The right column is a property of the request, not of the company.**

Summarising a public tender is not the same problem as reasoning over a client's
medical file, and the second does not become safe because procurement signed a
DPA. One organisation needs all three columns, chosen per s

项目描述

Annona — the sovereign execution kernel for AI agents. Decides where each step runs, enforces it, and records it.

相关仓库与替代方案

根据分类、Topic 和编程语言匹配的相似项目。

uzairansaruzi
精选
uzairansaruzi GitHub avatar

hermex

Hermex is a native SwiftUI iPhone app that lets you control a self-hosted Hermes AI agent directly from your phone, with no subscriptions, tracking, or third-party relay.

AI 与机器学习大语言模型
941
deerwork-ai
精选
deerwork-ai GitHub avatar

deer-workflow

An open-source Dynamic Workflow runtime that combines deterministic TypeScript orchestration with replaceable Agent runtimes.

AI 与机器学习大语言模型
312
makecindy
精选
makecindy GitHub avatar

cindy

Cindy is an open-source AI agent that runs locally on your machine, integrates multiple AI harnesses and models, and provides memory, skills, and automation to perform real work in your projects and apps.

AI 与机器学习大语言模型
958