Reading
jido_phx is a Phoenix LiveView application that demonstrates the capabilities of the Jido agent framework. It achieves real-time UI synchronization and LLM-driven human-AI collaborative workflows through a signal-driven agent architecture, providing an excellent example for building interactive AI applications.
Section 01
jido_phx is a Phoenix LiveView application that demonstrates the capabilities of the Jido agent framework. It achieves real-time UI synchronization and LLM-driven human-AI collaborative workflows through a signal-driven agent architecture, providing an excellent example for building interactive AI applications.
Section 02
Most current AI agents exist in the form of command lines or APIs. Users interact with them via text prompts, wait for responses, then proceed to the next round of dialogue. This model works for simple tasks but falls short for complex workflows—such as multi-step approvals or human-AI collaborative creation.
What we need is an interface that can reflect agent status in real time, support human-AI collaboration, and visualize workflows. This is exactly what the jido_phx project aims to demonstrate: how to deeply integrate AI agent frameworks with real-time web technologies.
Section 03
jido_phx is a Phoenix LiveView-based application that demonstrates the core capabilities of the Jido agent framework. It includes two main examples:
Section 04
| Layer | Technology Choice |
|---|---|
| Web Framework | Phoenix 1.8 + Phoenix LiveView 1.1 |
| Agent Framework | Jido 2.0 |
| LLM Client | req_llm 1.0 |
| Database | PostgreSQL via Ecto |
| Frontend Styling | Tailwind CSS |
| Build Tool | esbuild |
| HTTP Server | Bandit |
This technology stack reflects the modern best practices of the Elixir ecosystem.
Section 05
The core abstraction of the Jido framework is the "Signal". Agents do not directly respond to function calls; instead, they coordinate by sending and receiving signals. This model offers several advantages:
Section 06
Agents do not directly depend on each other; instead, they communicate via a signal bus. Adding new agents or modifying existing ones does not affect other components.
Section 07
All signal flows are traceable, facilitating debugging and auditing. The system can record the complete signal history and reproduce the agent decision-making process.
Section 08
Phoenix PubSub allows signal changes to be pushed to all connected clients instantly, enabling true real-time collaboration.
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.
libmlxforge is an embeddable MLX large language model (LLM) inference engine designed specifically for Apple Silicon. It provides a unified C ABI interface, supports calls from Node.js, Swift, and Rust, and features continuous batching, streaming output, JSON-constrained structured output, and embedding vector generation.