Reading
An open-source AI agent runtime layer that supports multi-provider agents, persistent context, relay execution, and deterministic workflows, enabling users to run durable AI agents in a self-hosted environment to operate local files, browsers, desktops, and workflows.
Section 01
An open-source AI agent runtime layer that supports multi-provider agents, persistent context, relay execution, and deterministic workflows, enabling users to run durable AI agents in a self-hosted environment to operate local files, browsers, desktops, and workflows.
Section 02
Section 03
PawFlow is a self-hosted AI agent runtime platform for real-world infrastructure. Positioned between chat agents, local tools, and production workflows, it provides persistent conversation, context, memory, files, processes, and provider session management. Through its relay architecture, PawFlow can execute file system, browser, desktop, and media tool operations on the user's local machine while maintaining the server's statelessness and security.
Section 04
Unlike many hosted AI coding assistants, PawFlow's core value proposition is "not handing over the workspace to a vendor-controlled agent cloud". Users fully own their data and execution environment, and can choose to deploy locally, in a private cloud, or in an offline environment. This design philosophy is particularly important for enterprises that value data privacy and compliance.
Section 05
PawFlow adopts a unique server-relay separation architecture. The server is responsible for hosting APIs, agent orchestration, process engines, and web interfaces, while the relay runs on the user's local machine (or in a Docker container) and executes actual tool operations via WebSocket connections. This means agents can operate the user's local codebase without the server needing direct access to user files, achieving a balance between functionality and security.
Section 06
The project emphasizes the concept of "durable"—conversations, shared context, agent-specific context, memory, knowledge graphs, diaries, project graphs, files, and buckets are all retained after a restart. This design allows AI agents to truly have long-term memory and context continuity, rather than starting from scratch every time.
Section 07
The server is the core of PawFlow, including the following subsystems:
Section 08
The relay is a key innovation in PawFlow's architecture. It runs on the user's machine and is responsible for executing over 90 tool operations:
Keep going with more reads from the same topic.
Nornir MCP Server is an enterprise-level server based on the Model Context Protocol (MCP). It seamlessly integrates large language models (such as Claude) with the Nornir network automation framework, supporting natural language orchestration for multi-vendor network devices (Cisco, Arista, Juniper, etc.), and providing production-grade features like a dual-engine architecture (NAPALM + Netmiko), intelligent filtering, and a secure sandbox.
Bibliothèque Française LLM is a structured indexing and annotation project for French public domain literature designed specifically for large language models (LLMs). It integrates multiple authoritative sources such as DraCor, Common Corpus, and Wikisource, providing metadata indexing categorized by genre, author, and era, as well as in-depth annotations for dramatic texts (including characters, lines, stage directions, etc.). Its aim is to enable LLMs to efficiently read and understand classic French literary works.
Splinter is a minimalist, high-performance key-value (KV) and vector storage system enabling zero-latency inter-process communication via shared memory and atomic operations. With only 766 lines of core code, it supports millions of operations per second and 768-dimensional vector storage, offering a new architectural approach for local LLM inference and data-intensive applications.
Folkering OS is the world's first AI-native bare-metal operating system, entirely written in Rust no_std without relying on Linux, POSIX, or libc. It can generate commands from scratch, compile them into WASM, and run them in 10 seconds, achieving true self-evolution.