Sidiora-Labs/paxeer-monorepo

Paxeer X is a Distributed HyperState Machine for payments, code execution and intent routing. Designed for Machines and the Users operating them.

What it solves

Paxeer X Network addresses the need for a blockchain system that can handle both standard EVM transactions and a separate execution environment for autonomous agents that require deterministic and verifiable outcomes. It provides a network where agent activity can be executed and accounted for in a deterministic manner, with signed receipts for each activity, ensuring trust and auditability.

How it works

The network consists of two execution domains: the Paxeer X chain (an EVM chain with BFT consensus) and the LayerX kernel (a C17-based deterministic execution environment for agents). The chain handles standard smart contract interactions, while LayerX executes agent programs and returns signed receipts. These receipts can be anchored to the chain via the layerxAnchor precompile. The monorepo includes the node, kernel, contracts, off-chain services (like agent daemon, MCP server), and client SDKs for various languages.

Who it’s for

Developers building decentralized applications, especially those involving autonomous agents or automated workflows that require deterministic execution and verifiable outcomes. Also for operators who want to run a node or kernel, and for users of the network's applications like perps trading, wallets, and launchpad.

Highlights

  • Dual-domain architecture: separates standard EVM execution from deterministic agent execution.
  • Signed receipts for agent activity, enabling trust and audit.
  • Broad SDK support (Go, Rust, Python, TypeScript, JVM, .NET, Swift).
  • Includes precompiles for many functions (staking, governance, IBC, etc.).
  • Off-chain services like agent daemon, MCP server, and interop gateway.
  • Planned features like perps, spot markets, and price feed.

Related

  • Project
    ratel-ai/ratel

    Ratel is a Rust‑backed retrieval engine with TypeScript and Python SDKs that lets LLM agents load only the tools and skill playbooks they actually need for each turn, cutting token cost and improving accuracy. It works with any model, requires no vector DB, and provides adapters for popular agent frameworks.

  • Project
    cv-cat/catbus

    A unified command-line interface that provides a consistent syntax for interacting with 11 different social media and web platforms, specifically designed to be easily integrated into AI agents.

  • Project
    StonyBrookNLP/appworld

    AppWorld is a high-fidelity execution environment simulating 9 everyday apps and ~100 simulated people, with 457 APIs, used to benchmark interactive coding and function-calling agents on complex real-world tasks. It provides task worlds, evaluation, and leaderboards for agent developers.

  • Project
    caura-ai/caura

    A governed shared-memory layer for AI agent fleets that allows multiple agents to store, recall, and compound knowledge across teams and organizations.