feat: ai agent blog post #1144
Draft
feat: ai agent blog post #1144
+148
−2
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Deploying windmilldocs with
|
| Latest commit: |
063c1c5
|
| Status: | ✅ Deploy successful! |
| Preview URL: | https://e42caec8.windmilldocs.pages.dev |
| Branch Preview URL: | https://fg-aiagentblog.windmilldocs.pages.dev |
hcourdent
reviewed
Nov 10, 2025
Contributor
hcourdent
left a comment
hcourdent
left a comment
There was a problem hiding this comment.
The videos are good but we can't see the characters, maybe we could try zooming by +10%
The first part looks very much like a text-from-bullet-points list, it would be great to add a little bit more consistence between the sections describing the features
Or maybe just do sub-sections for main features and Streaing, Memory etc. can be mentioned at the end of the big section. You kind of lose the reader imo
It lacks a lot of backlinks
blog/2025-11-10-ai-agents/index.mdx
Outdated
|
|
||
| The rise of large language models has created a new paradigm for workflow automation: instead of predefining every branch and decision in your workflows, you can let an AI agent reason about which tools to use and orchestrate them dynamically based on context. | ||
|
|
||
| AI agent steps in Windmill bring this capability to your workflows. Define the tools available, a Windmill script or any tools exposed by an MCP server. Then let the agent decide which to call, when to call them, and how to combine their results. The agent becomes a flexible orchestrator that adapts to each request rather than following a rigid script. |
Contributor
There was a problem hiding this comment.
backlink Script & Flows here, where it first appears
| Your browser does not support the video tag. | ||
| <a href={demoVideo}>Download the video</a>. | ||
| </video> | ||
|
|
Contributor
There was a problem hiding this comment.
give one intro sentence before jumping to sub-sections
blog/2025-11-10-ai-agents/index.mdx
Outdated
|
|
||
| ### Chat Mode | ||
|
|
||
| Enable Chat Mode in your flow, and Windmill automatically configures a conversational interface directly in the UI. Instead of traditional form inputs, users interact with your workflow through natural conversation. The agent maintains context across messages, allowing for back-and-forth exchanges where users can refine requests, ask follow-up questions, or provide additional information as the conversation evolves. |
blog/2025-11-10-ai-agents/index.mdx
Outdated
|
|
||
| ### Multi-provider support | ||
|
|
||
| Configure your AI agent with any provider: OpenAI, Anthropic, Azure OpenAI, Mistral, Google AI, Groq, Together AI, OpenRouter, or any custom or local endpoint you operate. |
Contributor
There was a problem hiding this comment.
backlink to the providers
blog/2025-11-10-ai-agents/index.mdx
Outdated
|
|
||
| ### Streaming | ||
|
|
||
| Long-running AI interactions can feel like black boxes. Streaming makes the agent's work visible in real-time. As the agent calls tools, processes their results, and formulates its response, users see updates showing exactly what's happening. This transparency is crucial for complex workflows where the agent might call multiple tools sequentially. |
blog/2025-11-10-ai-agents/index.mdx
Outdated
|
|
||
| ### Conversation memory | ||
|
|
||
| Agents can maintain conversation history, remembering earlier messages and building on previous exchanges. Configure how much context to preserve, and the agent will recall relevant information from earlier in the conversation. This enables truly conversational workflows where the agent doesn't treat each message in isolation but understands the broader context of what you're trying to accomplish. |
blog/2025-11-10-ai-agents/index.mdx
Outdated
|
|
||
| ### Structured output | ||
|
|
||
| Conversational text is useful, but sometimes you need data in a specific format that downstream systems can consume reliably. With JSON schema validation, you can ensure the AI's response conforms to a precise structure, returning a standardized object rather than free-form text. |
hugocasa
reviewed
Nov 11, 2025
blog/2025-11-10-ai-agents/index.mdx
Outdated
|
|
||
| ### Tool integration | ||
|
|
||
| Any Windmill [script](/docs/getting_started/scripts_quickstart) becomes a tool the AI agent can invoke. Write your tools in any of the 10+ languages Windmill supports—Python, TypeScript, Go, Rust, PHP, Bash, SQL, and more. The agent examines each tool's schema, understands its capabilities, and reasons about which tools to use based on the user's request. You can also use tools from the [Windmill Hub](https://hub.windmill.dev). |
Contributor
There was a problem hiding this comment.
you could add one sentence about how windmill scripts already have a schema which makes them perfect for tool definition
blog/2025-11-10-ai-agents/index.mdx
Outdated
|
|
||
|  | ||
|
|
||
| ### Conversational workflows |
Contributor
There was a problem hiding this comment.
before talking about chat mode it would be nice to say how you can interact with your ai agent steps using webhooks/http triggers (stream or not)
blog/2025-11-10-ai-agents/index.mdx
Outdated
|
|
||
| ### Multi-provider support and configuration | ||
|
|
||
| Configure your AI agent with any AI provider: [OpenAI](https://openai.com/), [Anthropic](https://www.anthropic.com/), [Azure OpenAI](https://learn.microsoft.com/en-us/azure/ai-services/openai/overview), [Mistral](https://mistral.ai/), [Google AI](https://ai.google.dev/models/gemini), [Groq](https://groq.com/), [Together AI](https://together.ai/), [OpenRouter](https://openrouter.ai/), or any custom or local endpoint you operate. |
Contributor
There was a problem hiding this comment.
blog/2025-11-10-ai-agents/index.mdx
Outdated
|
|
||
|  | ||
|
|
||
| **MCP integration**: Through [Model Context Protocol](https://modelcontextprotocol.io/docs/getting-started/intro) support, agents can connect to external MCP servers: file system browsers, database interfaces, API integrations, and custom business logic servers. This extends the agent beyond Windmill's internal capabilities to any MCP-compatible service. |
Contributor
There was a problem hiding this comment.
maybe move this to the tool section
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
No description provided.