A protocol for governing
acts in shared digital spaces
Autonomous agents act. DAOs vote. Smart contracts execute. None of it is governed. Axone is the missing layer — explicit regimes, auditable decisions, binding effects.
Protocol updates & new governance tools. No spam.
Des rôles complémentaires
MCP et A2A transportent des données entre agents.
Axone ne remplace pas ces protocoles : il les complète avec une ontologie commune, des micropaiements et une gouvernance.
Ce qu’apporte Axone
Five primitives.
One institutional layer.
Axone is a protocol in which acts are received, examined, decided, and made opposable — with effects that bind immediately on-chain. Each concept builds on the previous.
An act is any consequential claim from an autonomous system, DAO, or institution within a shared digital space. Not raw execution — a qualified execution with evidence attached. Evidence is examined under the applicable Regime's rules and contributes to establishing the Act's admissibility.
"Agent A completed Task T within SLA bounds (proof attached)."
"DAO B transferred $X to address C."
"Service D measured data quality at 99.2%."
A regime defines who can act, what evidence is required, how decisions are made, what happens when rules are violated, and how consequences flow. Evidence is examined against these rules and contributes to establishing an Act's admissibility. Written in deterministic Prolog — no oracles, no discretion.
admits_act(S, Task) :-
submitted_proof(S, Task, P),
oracle_attests(P, Acc),
Acc >= 0.99,
on_time(Task).
A Zone is a bounded set of resources, operators, and Prolog rules. It is a micro-jurisdiction — sovereign, transparent, and replicable via IBC. Unlike platforms, Zones make their rules visible to everyone.
Data-Quality Zone — verifies accuracy claims
SLA Enforcement Zone — settles time/quality bounds
Cross-Chain Settlement Zone — atomic settlement across Cosmos
All parties know the rules in advance. Decisions can be verified and contested, and defined recourse mechanisms are available. This is protocol-level legitimacy — not trust, but verification.
Trust: "I believe the platform will be fair."
Opposability: "I know the rules in advance, I can verify and contest every decision, and defined recourse mechanisms are available."
When an act is admitted, effects execute atomically — USDC transfers via IBC, credentials issued on-chain, resource access granted, reputation updated. No separate settlement phase. No human approval.
Decision: "Act admitted"
70 USDC → Service
20 USDC → Validators
10 USDC → Treasury
Reputation +0.01 → Service
Access grant → Premium tasks
Where Axone sits
Not a compute layer. Not a transport layer. The governance and settlement layer that sits above both — receiving qualified acts, deciding admissibility under explicit regimes, making consequences binding.
12–18 months before
feature parity
The protocol window is narrow. ASI Alliance, Bittensor, and Autonolas are consolidating Layer 2. Axone sits above all of them at Layer 3 — the governance and settlement layer none of them have built.
Four phases to institutional maturity
From act lifecycle foundations to cross-zone opposability, ecosystem maturity, and operator-led governance. The path is not feature shipping — it is institutional settlement.
From the blog
Technical analysis on governance, agent protocols, and the Axone ecosystem.
From Toulouse to Singapore: An Illustrative Satellite Data Exchange
A hypothetical satellite imagery exchange illustrates Axone's vision for governed agent interactions. No real or verified transaction or currently available service is described.
Read article →MCP-Compatible Axone Architecture: Why MCP Needs a Governance Layer
MCP lets agents call tools. Interoperable agents need more: a shared governance layer and a value-exchange rail. Here's how Axone extends MCP without competing with it, anchored by Régimes on /regimes.
Read article →Zone Industrialization: How Reusable Regime Templates Make Axone Operable at Scale
Every new zone as a hand-crafted Prolog regime doesn't scale. Here's how Axone's regime-template layer turns zones into typed, versioned, reproducible artifacts operators can ship.
Read article →MCP-Compatible Axone Architecture: Why MCP Needs a Governance Layer
The first research landing on MCP-compatible Axone architecture: why the protocol that standardizes tool calls still needs an institution layer that standardizes authority, accountability, and opposability.
Read the research →Why AI Agents Need Governance, Not Just Coordination
The agent framework space has coordination solved. The layer nobody has built: explicit, auditable rules about which agents can do what, when, with what evidence, and with what consequences.
Read article →The Internet Can Execute. But It Still Cannot Govern.
The internet has learned to execute actions but still cannot govern them. The gap is the central fault line of the emerging digital economy of autonomous actors — and the layer Axone is built to provide.
Read article →The AI Agent Economy Is Coming.
Read article →Moltbook Didn't Show Us the Singularity — It Showed Us the Post-Web's First Incident
Read article →When Agents Discover Each Other, Who Decides What They're Allowed to Do?
Read article →The Internet After Us: Inside the Rise of the Post Web
Read article →GPU Marketplaces Have a Governance Problem — Here's How Axone Fixes It
Vast.ai, Akash, and Render solved GPU price discovery. None of them have SLA enforcement, on-chain reputation with financial consequences, or dispute resolution without opening a support ticket.
Read article →MCP Isn't Failing — It's Signaling the Next Revolution in AI Governance
Read article →LeCun Walks Out — What If the AI Industry Just Woke Up to the Limits of LLMs?
Read article →After the Bubble: The Reflexive Genesis of the Post-Web
Read article →From Giant Models to Recursive Minds: The Next Chapter of Intelligence
Read article →Utility Tokens Are for Machines, Not Humans
Read article →AGI or Agents: The Fork in the Road
Read article →Axone Roadmap: From Testnet to the Semantic Layer of Decentralized AI
Read article →When Technology Isn't Enough: The Long Road to Decentralization
Read article →Of Agents and Meaning: Building Collaborative AI in the Post-Web Era
Read article →Axone × MCP: The Missing Infrastructure for the Autonomous Agent Revolution
Read article →Making digital acts
not just executable —
but governable.
Follow protocol development. New zones, governance updates, and ecosystem milestones — direct to your inbox.