Top news
Investing in the public benefit OfficialAnthropic NewsOct 18Popular Musk Pledges SpaceX Shares to ‘Every Kid in America’ WebBloomberg TechnologyOct 19Popular Websites are closing their doors to AI agents WebThe Verge AIOct 17 This is the new ChatGPT Voice, powered by GPT Live OfficialOpenAI on YouTubeOct 18 How AI giants will reshape global philanthropy WebAxios AIOct 19 Gemini 4 Argon Really Surprised Me WebTwo Minute PapersOct 19 Google Gemini 4 agents just beat Minecraft with no human intervention WebEngadget AIOct 19 Who could it be? 🧐 OfficialGoogle on YouTubeOct 18 Anthropic gives outside AI watchdog “employee-like” access WebArs Technica AIOct 18 What are self-replicating ‘AI worms’, and why is Anthropic worried about them? WebThe Guardian AIOct 18
GlossaryAgentsIntermediate

Agent workflows

Also written: Agent-led workflows

Automated sequences of actions by AI agents to achieve goals, often using tools and decision making.

In plain terms

Agent workflows are like a to-do list for an AI. Instead of just answering a single question, the AI can follow a series of steps to complete a larger task.

Think of it like asking an AI to plan a trip. It might first search for flights, then look for hotels, and finally book both. Each of these is a step in the agent workflow, and the AI uses different tools (like a flight search engine or a booking website) to complete them.

The expert version

Agent workflows represent automated sequences of actions executed by AI agents to accomplish specific objectives. These workflows typically involve a combination of reasoning, decision-making, and the utilization of external tools or APIs. The agent plans a series of steps, executes them, and uses the results to inform subsequent actions, enabling complex task completion beyond single-turn interactions.

Key components include planning modules, tool-use interfaces, and execution engines. Tradeoffs involve increased complexity and potential for error propagation versus enhanced capability and automation of multi-step processes.

Why it matters now

Frontier AI labs are focusing on agent workflows as they develop more capable AI models. These models, like Google's Gemini Pro, are being enhanced with better reasoning and tool use to enable more complex, automated task completion. This allows AI agents to perform multi-step processes autonomously, moving beyond single-task execution.

Where it shows up 11 items, newest first
How each company uses it
Google Gemini
Google Gemini is highlighting its Pro model's enhanced reasoning and custom tool support for developers building complex agent workflows. They also mention their Flash-Lite model for high-volume tasks, implying agent workflows can be optimized for efficiency.
Microsoft Copilot
Microsoft Copilot is described as orchestrating AI across media workflows, suggesting it manages or enables agent-led workflows for content creation.
OpenAI ChatGPT
OpenAI's ChatGPT is mentioned in relation to agent workflows, indicating its use in sequences of automated actions for specific goals.
Related terms

AI agentsTool useOrchestrationPrompt engineeringAutonomous systemsLLM applications

Often confused with
Chaining
Chaining involves linking multiple LLM calls sequentially, but agent w
Task automation
Task automation is a general concept; agent workflows are a spe
Function calling
Function calling allows an LLM to call a specific function, wh
Origin

The concept of agent workflows evolved from research into intelligent agents and autonomous systems, with significant recent development driven by the capabilities of large language models.

Try this about 5 minutes

Try using a chatbot that can browse the web. Ask it to find the current weather in London and then tell you what to wear. This simple task involves the AI using a search tool (weather website) and then using that information to provide a recommendation, demonstrating a basic agent workflow.

Try it: Playbook entries that use thisPlaybook →

The Playbook: reusable how-tos, prompts, configs and tips pulled out of the news.

Questions people ask

What makes an AI workflow an 'agent' workflow?
An agent workflow is characterized by an AI agent that can plan, make decisions, and use tools to complete a series of tasks autonomously towards a goal.
How do agent workflows differ from simple scripts?
Agent workflows are more dynamic. They can adapt their steps based on intermediate results and make decisions, whereas scripts typically follow a fixed, predetermined sequence of commands.

First seen on Subvolts Mar 6, 2026. Updated Oct 15, 2026. Written by an AI model from the items above and reviewed against them weekly; report an error.