Reading
A zero-dependency TypeScript CLI runtime that allows dynamic workflow scripts originally only executable within Claude Code to run cross-platform, supporting various coding agent CLIs like Codex, Gemini, Qwen, Kimi, etc.
Section 01
A zero-dependency TypeScript CLI runtime that allows dynamic workflow scripts originally only executable within Claude Code to run cross-platform, supporting various coding agent CLIs like Codex, Gemini, Qwen, Kimi, etc.
Section 02
Section 03
The dynamic workflows feature introduced by Claude Code allows developers to orchestrate multiple coding agents using JavaScript scripts, enabling complex task decomposition and parallel processing. However, these workflow scripts can only run in Claude Code's private runtime, leading to a severe platform lock-in problem.
When you spend time crafting a sophisticated 'in-depth research' workflow—one that launches multiple web searches in parallel, performs adversarial fact-checking, and finally synthesizes a report with cited sources—you cannot migrate this workflow to other coding agents. This fragmentation impedes workflow reuse and ecosystem growth.
Section 04
Open Dynamic Workflows (ODW) is a TypeScript/Node CLI runtime specifically designed to solve the above problem. It allows dynamic workflow scripts originally only executable within Claude Code to run on any coding agent CLI, including Codex, Claude Code, Gemini, Qwen, Kimi, and even custom CLIs.
Section 05
True Portability: The same workflow script can switch seamlessly between different agents by simply changing the adapter configuration.
Preserve Claude Code Dialect: Fully compatible with Claude Code's workflow syntax—export const meta, injected global variables like agent/parallel/pipeline/phase/log/args/budget, and supports top-level await and return.
Context-Free, Scalable Execution: Workflow plans are defined in code; intermediate work does not pollute the main session context, allowing dozens of sub-agents to be launched in parallel.
Reliable Result Delivery: Supports JSON Schema structured output, with validation and retry mechanisms to ensure reliable composition of multi-stage pipelines.
Background Execution & Observability: Each run is an independent background process with a dedicated runtime directory, supporting commands like status, logs --follow, result, pause/stop, etc.
Zero Runtime Dependencies: The engine is based on pure asynchronous TypeScript; parallel is simply Promise.all, and workflow scripts remain in pure .js format.
Section 06
ODW uses a clear layered architecture to isolate complexity into different levels:
Section 07
The src/adapters/ directory contains a unified CLI call abstraction, handling configuration parsing, placeholder replacement, runner implementation, and built-in adapters for Codex, Claude Code, Gemini, Qwen, and Kimi.
Section 08
src/bridge.ts converts a single agent() call into a CLI run, handling schema validation and retry logic.
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.