Tag
This article discusses the growing need for multi-agent orchestration in AI coding workflows, highlighting several tools like AutoGen, Claude Code, and CrewAI that enable agents to collaborate, coordinate, and share information to handle complex tasks beyond manual management.
The article discusses how AI transformation in the enterprise requires changing underlying workflows and deploying agents against business processes, rather than just rolling out tools to end users. It emphasizes deep domain expertise, data organization, and comprehensive evaluations for ROI.
A community discussion asking for real-world, production-grade automations, AI agents, or feedback loops that solve day-to-day problems, excluding one-off prototypes.
Noofy is an open-source app that simplifies complex ComfyUI workflows by turning them into easy-to-use dashboards.
An open-source workspace designed for AI agents and workflows, facilitating development and integration.
Peter Gostev compares Fable and GPT-5.6-Sol, describing Fable as more intelligent and eloquent but less reliable, while GPT-5.6-Sol is a diligent workhorse that excels at executing tasks and maintaining code patterns. He provides detailed observations on UI, writing, robustness, and other features.
Swarm is a Swift framework for building agentic workflows and multi-agent systems, with type-safe tool calling, durable checkpointing, and support for multiple LLM providers.
GitHub introduces custom agents in Copilot CLI, allowing developers to define specialized, reusable workflows in Markdown files for tasks like security audits and release notes.
A guide to using Fable 5 and Claude Code to create automated workflows (loops and goals) for AI agents, including 25 workflow examples with prompts and tool integration.
Forward Future has open-sourced a library of 70 AI agent workflows with copy-paste prompts and defined stopping conditions, covering engineering, evaluation, operations, and content tasks.
A researcher and engineer discusses the benefits of multimodal prompting for AI agents, explaining how combining voice, screen annotations, and actions improves agent performance and reduces frustration.
ComfyClaw is an agentic skill evolution framework for ComfyUI image generation workflows, using typed graph editing and region-level VLM verifiers to translate visual failures into repair suggestions, outperforming baselines across multiple configurations.
This article explains how using Postgres transactions for workflow state co-located with application data eliminates idempotency and atomicity problems in distributed workflows, providing exactly-once execution.
LangChain's Deep Agents now supports dynamic subagents, where the main agent writes orchestration code to coordinate work at scale, enabling workflows like processing hundreds of documents with deterministic coverage.
A tweet thread discusses best practices for AI token cost optimization, arguing that a deep understanding of workflows and architecture is needed for enterprises to maximize ROI, and that this represents a major opportunity for applied AI companies.
A discussion on when to use scripts versus AI agents, arguing that most claimed AI agents are actually simple scripts, and only truly dynamic decision-making tasks warrant agents.
TUA-Bench is a comprehensive benchmark for evaluating general-purpose terminal-use agents across diverse digital activities and specialized workflows, revealing significant performance gaps among current frontier agents.
The author observes that Zapier handles fixed workflows well, but variable workflows are where they want to use an AI agent.
Two senior Google Cloud engineers gave a live lecture on building multi-agent workflows, discussing Google's open-sourced production multi-agent system.
A detailed guide covering 50 advanced Obsidian workflows, plugins, and setups, including Dataview queries, Templater automation, and Canvas spatial thinking techniques, aimed at transforming Obsidian into a personal operating system.