docs: agent blocks by mendonk · Pull Request #11529 · langflow-ai/langflow

mendonk · 2026-02-02T14:25:55Z

Add a quickstart, component docs, and release note for the Agent Blocks feature.

coderabbitai · 2026-02-02T14:26:08Z

Important

Review skipped

Auto reviews are disabled on base/target branches other than the default branch.

Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

🔍 Trigger a full review

✨ Finishing touches

🧪 Generate unit tests (beta)

Create PR with unit tests
Post copyable unit tests in a comment
Commit unit tests in branch docs-agent-blocks

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

…ngflow into docs-agent-blocks

mendonk · 2026-02-04T17:18:23Z

            "Components/components-bundles",
+            {
+              type: "category",
+              label: "Agent Blocks",


Agent blocks

aimurphy

Approved but please see my comments

aimurphy · 2026-02-04T19:16:34Z

@@ -0,0 +1,34 @@
+---
+title: AgentLoop


The titles of these pages don't match the spelling used in the screenshot and on the pages. For example AgentLoop vs Agent Loop.

The "no space" versions are also in other places, like AgentLoop parameters.

aimurphy · 2026-02-04T19:18:00Z

+
+An agent is fundamentally a while loop.
+The while loop accepts input and iterates over the agent's reasoning.
+The agent's reasoning sends messages to an LLM and examines the LLM's response to determine where to route the request.


Suggested change

The agent's reasoning sends messages to an LLM and examines the LLM's response to determine where to route the request.

The agent's reasoning sends messages to an LLM, and then it examines the LLM's response to determine where to route the request.

aimurphy · 2026-02-04T19:20:37Z

+```
+ChatInput → WhileLoop → AgentStep → [Tool Calls] → ExecuteTool → WhileLoop
+                              ↓ [AI Message - done]
+                         ChatOutput
+```


I suggest a mermaid diagram here instead.
You can use the free editor https://mermaid.live/edit

I can't remember if mermaid was already enabled in this repo https://docusaurus.io/docs/next/markdown-features/diagrams

aimurphy · 2026-02-04T19:21:34Z

+                         ChatOutput
+```
+
+Langflow's [**Agent** component](/components-agents) is composed of this same architecture, and is suitable for most tasks that require a reasoning loop that can choose from a list of registered functions.


Suggested change

Langflow's [**Agent** component](/components-agents) is composed of this same architecture, and is suitable for most tasks that require a reasoning loop that can choose from a list of registered functions.

Langflow's [**Agent** component](/components-agents) uses this architecture, and it is suitable for most tasks that require a reasoning loop that can choose from a list of registered functions.

aimurphy · 2026-02-04T19:24:06Z

+But what if you want to modify the agent's behavior? For example, how do you include a validation step in your tool outputs before execution?
+
+Use **Agent Blocks** to construct an agent in Langflow's visual builder.


Suggested change

But what if you want to modify the agent's behavior? For example, how do you include a validation step in your tool outputs before execution?

Use **Agent Blocks** to construct an agent in Langflow's visual builder.

If you want to modify the agent's behavior, you can use **Agent Blocks** to construct an agent. For example, you could add a validation step to `tool_calls` outputs.

aimurphy · 2026-02-04T20:03:22Z

+import Icon from "@site/src/components/icon";
+import PartialParams from '@site/docs/_partial-hidden-params.mdx';
+
+The **Think Tool** component provides think tool support for step-by-step reasoning. It enables agents to perform explicit reasoning steps before taking action, improving the quality of tool selection and parameter generation.


The first sentence is repetitive and doesn't help explain what it does.

aimurphy · 2026-02-04T20:03:51Z

+**Think Tool** is typically used as a tool that the [**Agent Step** component](/agent-blocks-agent-step) can call during reasoning:
+
+```
+AgentStep → [Think Tool Call] → ThinkTool → [Reasoning Output] → AgentStep


I'm not sure this is adding any value.

aimurphy · 2026-02-04T20:05:33Z

+| chat_history | Chat Memory | Data[] | Input parameter. Chat history for context-aware responses. |
+| input_value | Input | Message | Input parameter. The input message to process. |
+| tool_calls | Tool Calls | ToolCall[] | Output parameter. Tool calls requested by the LLM (routes to [**Execute Tool** component](/agent-blocks-execute-tool)). |
+| ai_message | AI Message | Message | Output parameter. Final answer from the LLM (routes to [**Chat Output** component](/chat-input-and-output)). |


On the quickstart, the outputs are Response and Tool Calls.

aimurphy · 2026-02-04T20:07:46Z

+import PartialParams from '@site/docs/_partial-hidden-params.mdx';
+
+The **Agent Loop** component is a pre-composed component that encapsulates the complete agent loop. It combines the [**While Loop** component](/agent-blocks-while-loop), [**Agent Step** component](/agent-blocks-agent-step), and [**Execute Tool** component](/agent-blocks-execute-tool) into a single, ready-to-use component.
+


I would add an explanation or list here about when to use:

Individual Agent Blocks components: You want to build a custom agent with extra steps or logic beyond regular tool calls.

Agent Loop component: ?

Agents component: ?

aimurphy · 2026-02-04T20:08:29Z

          label: "Bundles",
          items: [
            "Components/components-bundles",
+            {


strange this is under bundles instead of core 🤷🏻‍♀️

github-actions · 2026-02-04T22:05:39Z

Build successful! ✅
Deploying docs draft.
Deploy successful! View draft

mendonk · 2026-02-06T19:20:47Z

Closing as won't do for 1.8 release.

mendonk added 5 commits January 26, 2026 14:49

introduction

21bc033

initial-content-no-quickstart

16b885d

Merge branch 'main' into docs-agent-blocks

98e63dc

Merge branch 'main' into docs-agent-blocks

93235ff

Merge branch 'main' into docs-agent-blocks

8ef28a2

mendonk self-assigned this Feb 2, 2026

Merge branch 'docs-1.8-release' into docs-agent-blocks

2212944

github-actions Bot added documentation Improvements or additions to documentation and removed documentation Improvements or additions to documentation labels Feb 2, 2026

This comment has been minimized.

Sign in to view

mendonk added 2 commits February 2, 2026 11:43

place-content-under-bundles

d9fc5b2

Merge branch 'docs-agent-blocks' of https://github.com/langflow-ai/la…

53e646a

…ngflow into docs-agent-blocks

github-actions Bot added documentation Improvements or additions to documentation and removed documentation Improvements or additions to documentation labels Feb 2, 2026

This comment has been minimized.

Sign in to view

mendonk mentioned this pull request Feb 3, 2026

Docs: 1.8 release #11295

Merged

mendonk added 2 commits February 3, 2026 14:56

Merge branch 'main' into docs-agent-blocks

9f8f4e1

add-quickstart

708d770