Annona is an open-source 'sovereign execution kernel' that decides, enforces, and records where each step of an AI agent run may execute, with local-by-default and default-deny policy.

Stars

11

7-day growth

No data

Forks

1

Open issues

0

License

Apache-2.0

Last updated

2026-07-31

AI repository intelligence
FR-AI / ANALYSIS

Why it is worth attention

It reframes AI deployment as per-request policy rather than a single architecture choice, enforces placement in the runtime with fail-closed semantics, and makes the decision verifiable via a hash-chained ledger.

Who it is for

  • Organizations handling sensitive client data and needing GDPR/EU AI Act alignment
  • Teams deploying AI agents across local, EU-cloud, and frontier APIs
  • Platform/security engineers who need auditable AI runtime policy
  • Legal/professional services firms with confidential documents

Use cases

  • Running agent steps on local models when they touch restricted files, while allowing public tasks to use cheaper frontier APIs
  • Holding a step instead of rerouting when no permitted substrate is available
  • Importing Claude Skills and pinning them to run only locally
  • Redacting restricted identifiers locally before sending reclassified content to a frontier model

Strengths

  • Fail-closed, default-deny policy with monotone contamination tracking and no silent downgrade
  • Hash-chained ledger with verify/why/audit commands and a nine-check acceptance run
  • One binary/one release supports detached, attached, and appliance topologies without forks
  • Layered architecture with enforced build-time contracts preventing decision-layer shortcuts to provider adapters

Considerations

  • Explicit gaps remain: no grammar-constrained tool calls, no external ledger anchor, no automatic resume of queued steps
  • Legacy config path remains allow-by-default until annona policy init is run
  • Unsigned beta desktop installers require manual OS overrides

README quick start

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

Description

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

Related repositories

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

uzairansaruzi
Featured
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 & Machine LearningLarge Language Models
941
deerwork-ai
Featured
deerwork-ai GitHub avatar

deer-workflow

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

AI & Machine LearningLarge Language Models
312
makecindy
Featured
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 & Machine LearningLarge Language Models
958