change langchain quickstart by npentrel · Pull Request #3531 · langchain-ai/docs

Naomi Pentrel (npentrel) · 2026-04-10T13:04:12Z

I've changed the langchain quickstart:
it installs deepagents, explains the difference and points people towards using deepagents for real-world applications.
it now also includes more on tracing and more models

https://langchain-5e9cc07a-preview-naomid-1776417660-b4d4d36.mintlify.app/oss/python/langchain/quickstart

github-actions · 2026-04-10T13:05:59Z

Mintlify preview ID generated: preview-naomid-1775826331-671d609

github-actions · 2026-04-10T16:17:08Z

Mintlify preview ID generated: preview-naomid-1775837804-02292a4

github-actions · 2026-04-10T16:38:45Z

Mintlify preview ID generated: preview-naomid-1775839101-6e570c5

Mason Daugherty (mdrxy) · 2026-04-12T17:06:25Z

-* Set the `ANTHROPIC_API_KEY` environment variable in your terminal
+:::python
+<CodeGroup>
+    ```bash pip


nit: i think we should begin to show uv instead of pip but that's open for debate

Mason Daugherty (mdrxy) · 2026-04-12T17:07:12Z

+    ```bash HuggingFace
+    export HUGGINGFACEHUB_API_TOKEN="hf_..."
+    export TAVILY_API_KEY="your-tavily-api-key"
+    ```


nit: we should add an other tab

Mason Daugherty (mdrxy) · 2026-04-12T17:08:11Z

+        <Tab title="LangChain agents">
+
+        :::python
+        ```python wrap


nit: can we make this an expand code block? really long without adding much

Mason Daugherty (mdrxy) · 2026-04-12T17:08:26Z

+Continue with:
+
+- **LangChain agents**: [Add and manage memory](/oss/langgraph/add-memory#manage-short-term-memory), [deploy to production](/oss/langgraph/deploy)
+- **Deep agents**: [Customization options](/oss/deepagents/customization), [persistent memory](/oss/deepagents/long-term-memory), [deploy to production](/oss/langgraph/deploy)


Suggested change

- **Deep agents**: [Customization options](/oss/deepagents/customization), [persistent memory](/oss/deepagents/long-term-memory), [deploy to production](/oss/langgraph/deploy)

- **Deep Agents**: [Customization options](/oss/deepagents/customization), [persistent memory](/oss/deepagents/long-term-memory), [deploy to production](/oss/langgraph/deploy)

github-actions · 2026-04-15T15:28:22Z

Mintlify preview ID generated: preview-naomid-1776266869-edfb3a1

Naomi Pentrel (npentrel) · 2026-04-16T11:18:42Z

+        [
+            {
+                index: 0,
+                type: 'text',
+                text: 'I now have the full text. Let me carefully analyze it for both counts.\n' +
+                '\n' +
+                '---\n' +
+                '\n' +
+                '## Analysis\n' +
+                '\n' +
+                '### Question 1 — Lines containing `Elizabeth`\n' +
+                '\n' +
+                'The tool returns the file as one large block of text. I need to count **lines** (newline-delimited) that contain the substring `Elizabeth` at least once.\n' +
+                '\n' +
+                'Working through the text systematically, I can identify every line that contains `Elizabeth`. Rather than risk an off-by-one manual count across thousands of lines without a programmatic counter, I need to be transparent: **the tool loads the document as a single text blob**, and I have no line-numbering facility available to me. I can read the text and reason about it, but I cannot execute a `grep -c` or equivalent command. Attempting to hand-count every occurrence across ~13,000+ lines of text would be error-prone and likely to produce a fabricated number.\n' +
+                '\n' +
+                '### Question 2 — First line containing `Darcy`\n' +
+                '\n' +
+                'Scanning from the very top of the file, the first occurrence of `Darcy` appears in the **Preface** by George Saintsbury. The relevant passage reads:\n' +
+                '\n' +
+                `> *"…the false account of Darcy's conduct given by Wickham…"*\n` +
+                '\n' +
+                'That line is deep in the Preface section. The very first line in the file containing `Darcy` is:\n' +
+                '\n' +
+                '> `"…the false account of Darcy\'s conduct given by Wickham, and by the awkwardness…"`\n' +
+                '\n' +
+                'Again, without a line counter I cannot give a verified 1-based line number.\n' +
+                '\n' +
+                '---\n' +
+                '\n' +
+                '## Results\n' +
+                '\n' +
+                '```json\n' +
+                '{\n' +
+                '  "question_1": {\n' +
+                '    "answer": null,\n' +
+                `    "how_you_computed_counts": "The fetch_text_from_url tool returns the document as a single unstructured text blob with no line numbers. I have no available tool to execute grep, wc, or any line-counting operation. Manually counting every newline-delimited line containing 'Elizabeth' across ~13,000+ lines would be unreliable and likely to produce a fabricated figure. Therefore I return null rather than guess."\n` +
+                '  },\n' +
+                '  "question_2": {\n' +
+                '    "answer": null,\n' +
+                `    "how_you_computed_counts": "The first occurrence of 'Darcy' in the file is in the Preface (George Saintsbury's essay), in the sentence '…the false account of Darcy's conduct given by Wickham…'. However, without a line-numbering facility I cannot verify the exact 1-based line number. The tool provides no line index, and manually counting blank lines, illustration captions, and wrapped prose lines to arrive at a precise number would be error-prone. I return null rather than fabricate a line number."\n` +
+                '  },\n' +
+                '  "synopsis": "Pride and Prejudice (1813) by Jane Austen follows Elizabeth Bennet, the second of five daughters in a genteel but financially precarious English family, as she navigates questions of marriage, social class, and personal integrity in early nineteenth-century rural England. The novel centers on her evolving relationship with the wealthy and initially aloof Fitzwilliam Darcy, whose pride and her own prejudice must each be overcome before the two can recognize their mutual regard and agree to marry."\n' +
+                '}\n' +
+                '```\n' +
+                '\n' +
+                '### Summary\n' +
+                '\n' +
+                '| Field | Value |\n' +
+                '|---|---|\n' +
+                '| **Lines containing `Elizabeth`** | `null` — the tool returns raw text with no line-number index; a programmatic counter (e.g., `grep -c`) is required for a verified answer |\n' +
+                '| **First line number with `Darcy`** | `null` — same limitation; the word first appears in the Preface in the phrase *"Darcy\'s conduct given by Wickham"*, but the exact 1-based line number cannot be verified without a line counter |\n' +
+                '| **Synopsis** | *Pride and Prejudice* follows Elizabeth Bennet as she navigates marriage prospects and social pressures in Regency England, ultimately overcoming mutual pride and prejudice to unite with the wealthy Mr. Darcy. |'
+            }
+        ]


Suggested change

[

{

index: 0,

type: 'text',

text: 'I now have the full text. Let me carefully analyze it for both counts.\n' +

'\n' +

'---\n' +

'\n' +

'## Analysis\n' +

'\n' +

'### Question 1 — Lines containing `Elizabeth`\n' +

'\n' +

'The tool returns the file as one large block of text. I need to count **lines** (newline-delimited) that contain the substring `Elizabeth` at least once.\n' +

'\n' +

'Working through the text systematically, I can identify every line that contains `Elizabeth`. Rather than risk an off-by-one manual count across thousands of lines without a programmatic counter, I need to be transparent: **the tool loads the document as a single text blob**, and I have no line-numbering facility available to me. I can read the text and reason about it, but I cannot execute a `grep -c` or equivalent command. Attempting to hand-count every occurrence across ~13,000+ lines of text would be error-prone and likely to produce a fabricated number.\n' +

'\n' +

'### Question 2 — First line containing `Darcy`\n' +

'\n' +

'Scanning from the very top of the file, the first occurrence of `Darcy` appears in the **Preface** by George Saintsbury. The relevant passage reads:\n' +

'\n' +

`> *"…the false account of Darcy's conduct given by Wickham…"*\n` +

'\n' +

'That line is deep in the Preface section. The very first line in the file containing `Darcy` is:\n' +

'\n' +

'> `"…the false account of Darcy\'s conduct given by Wickham, and by the awkwardness…"`\n' +

'\n' +

'Again, without a line counter I cannot give a verified 1-based line number.\n' +

'\n' +

'---\n' +

'\n' +

'## Results\n' +

'\n' +

'```json\n' +

'{\n' +

' "question_1": {\n' +

' "answer": null,\n' +

` "how_you_computed_counts": "The fetch_text_from_url tool returns the document as a single unstructured text blob with no line numbers. I have no available tool to execute grep, wc, or any line-counting operation. Manually counting every newline-delimited line containing 'Elizabeth' across ~13,000+ lines would be unreliable and likely to produce a fabricated figure. Therefore I return null rather than guess."\n` +

' },\n' +

' "question_2": {\n' +

' "answer": null,\n' +

` "how_you_computed_counts": "The first occurrence of 'Darcy' in the file is in the Preface (George Saintsbury's essay), in the sentence '…the false account of Darcy's conduct given by Wickham…'. However, without a line-numbering facility I cannot verify the exact 1-based line number. The tool provides no line index, and manually counting blank lines, illustration captions, and wrapped prose lines to arrive at a precise number would be error-prone. I return null rather than fabricate a line number."\n` +

' },\n' +

' "synopsis": "Pride and Prejudice (1813) by Jane Austen follows Elizabeth Bennet, the second of five daughters in a genteel but financially precarious English family, as she navigates questions of marriage, social class, and personal integrity in early nineteenth-century rural England. The novel centers on her evolving relationship with the wealthy and initially aloof Fitzwilliam Darcy, whose pride and her own prejudice must each be overcome before the two can recognize their mutual regard and agree to marry."\n' +

'}\n' +

'```\n' +

'\n' +

'### Summary\n' +

'\n' +

'| Field | Value |\n' +

'|---|---|\n' +

'| **Lines containing `Elizabeth`** | `null` — the tool returns raw text with no line-number index; a programmatic counter (e.g., `grep -c`) is required for a verified answer |\n' +

'| **First line number with `Darcy`** | `null` — same limitation; the word first appears in the Preface in the phrase *"Darcy\'s conduct given by Wickham"*, but the exact 1-based line number cannot be verified without a line counter |\n' +

'| **Synopsis** | *Pride and Prejudice* follows Elizabeth Bennet as she navigates marriage prospects and social pressures in Regency England, ultimately overcoming mutual pride and prejudice to unite with the wealthy Mr. Darcy. |'

}

]

{

"lines_containing_Elizabeth": null,

"first_line_number_containing_Darcy": null,

"synopsis": "Pride and Prejudice follows Elizabeth Bennet as she navigates family pressures, social expectations, and evolving relationships within the landed gentry of early nineteenth-century England. The novel centers on misunderstandings, judgments, and personal growth, especially in the developing connection between Elizabeth and Mr. Darcy.",

"how_you_computed_counts": "I attempted to load the Project Gutenberg plain-text file from the provided URL using the available document-fetch tool, but the fetch failed. Because I could not retrieve and inspect the actual file contents, I could not verify either the number of lines containing 'Elizabeth' or the 1-based line number of the first line containing 'Darcy', so I returned null for both rather than guessing."

}

github-actions · 2026-04-16T11:21:40Z

Mintlify preview ID generated: preview-naomid-1776338473-cf697ea

Lauren Hirata Singh (lnhsingh) · 2026-04-16T15:52:16Z

love the example

github-actions · 2026-04-16T17:49:53Z

❌ Import check failed

This PR contains imports from langchain_core that should be imported from langchain instead.

Detailed issues

Analyzing diff for import issues...
 Found 1 import issues:

File: src/code-samples/deepagents/skills-sandbox.py
Line: 13
Issue: Import from langchain.chat_models instead of langchain_core.language_models
Current:   from langchain_core.language_models import BaseChatModel
Suggested: from langchain.chat_models import BaseChatModel
--------------------------------------------------------------------------------

 Found 1 import issues that need to be fixed.

Why this is a problem

The langchain package re-exports many modules and classes from langchain_core. When possible, imports should use langchain instead of langchain_core for:

Better user experience (single import source)
Consistency across documentation
Reduced cognitive load for users

How to fix

Replace the imports as suggested above. For example:

❌ from langchain_core.messages import HumanMessage
✅ from langchain.messages import HumanMessage

🤖 Automated check

This check is based on the latest analysis of langchain re-exports from langchain_core.

github-actions · 2026-04-16T17:50:57Z

Mintlify preview ID generated: preview-naomid-1776361822-1e67918

github-actions · 2026-04-16T17:54:54Z

Mintlify preview ID generated: preview-naomid-1776362063-a92ddda

github-actions · 2026-04-17T09:21:29Z

Mintlify preview ID generated: preview-naomid-1776417660-b4d4d36

Mason Daugherty (mdrxy) · 2026-04-17T16:01:11Z

    2. A standard message content format: Model APIs evolved from returning messages with a simple content string to more complex output types - reasoning blocks, citations, server-side tool calls, etc. LangChain evolved its message formats to standardize these across providers.
 </Update>
+
+<Update label="2026-03-15">


This date is incorrect;

0.0.1 was published Jul 29, 2025

I based this on when we announced it to the world. Though it's also possible we don't want this here at all until v1.0. vivek (@vtrivedy) what do you think?

Mason Daugherty (mdrxy) · 2026-04-17T16:02:12Z

+<Update label="2026-03-15">
+    **Deep Agents is released** as an open-source agent harness built on LangGraph.
+
+    While LangChain provides flexible building blocks for custom agent architectures, [Deep Agents](/oss/langchain/overview) offers a batteries-included option for complex, long-running tasks like research and coding. It adds built-in planning tools, a virtual filesystem with pluggable backends (in-memory, disk, LangGraph store, sandboxes), subagent spawning for context isolation, and a terminal CLI. Use Deep Agents for more autonomous agents with predefined tools; use LangChain for full control over your agent architecture.


Not sure we should mention CLI

Mason Daugherty (mdrxy) · 2026-04-17T16:05:33Z

 </Tip>

-## Requirements
+## Install dependencies


this makes me think; do we need a dedicated install page? https://docs.langchain.com/oss/python/langchain/install

can we fold these together or remove the install section from this page?

let's do that separately. This PR is opening a big enough can of worms as it is :)

Mason Daugherty (mdrxy) · 2026-04-17T16:06:46Z

+    # Requires Python 3.10+
+    ```
+
+    ```bash pip


if we're going to show pip we should show steps for creating a venv (same as we do for uv)

Mason Daugherty (mdrxy) · 2026-04-17T16:07:25Z

+    uv init
+    uv add langchain deepagents
+    uv sync
+    # Requires Python 3.10+


idk if we need these python version comments any more

Mason Daugherty (mdrxy) · 2026-04-17T16:09:01Z

+    result = agent.invoke(
+        {"messages": [{"role": "user", "content": "What's the weather in San Francisco?"}]}
+    )
+    print(result["messages"][-1].content)


We should be using .content_blocks

Mason Daugherty (mdrxy) · 2026-04-17T16:12:26Z

 :::

+When you run the code and prompt the agent to tell you about the weather in San Francisco, the agent uses that input and its available context.
+The agent understands that you are asking about the weather for the city San Francisco and therefore calls the weather tool with the provided city name. This is how you get the response `"It's always sunny in San Francisco!"`.


we mention the response but never show it (i don't think we need the final sentence)

Mason Daugherty (mdrxy) · 2026-04-17T16:14:14Z

+When you run the code and prompt the agent to tell you about the weather in San Francisco, the agent uses that input and its available context.
+The agent understands that you are asking about the weather for the city San Francisco and therefore calls the weather tool with the provided city name. This is how you get the response `"It's always sunny in San Francisco!"`.
+
 <Tip>


i think we should just fold this into the "other" tab

if we're keeping it i think it should be less ambiguous

Mason Daugherty (mdrxy) · 2026-04-17T16:15:23Z

+
+        Now assemble your agent with all the components and run it.
+
+        There are two different frameworks for creating agents: LangChain agents and deep agents.


should always be capitalized right?

Suggested change

There are two different frameworks for creating agents: LangChain agents and deep agents.

There are two different frameworks for creating agents: LangChain agents and Deep Agents.

Naomi Pentrel (npentrel) requested a review from Lauren Hirata Singh (lnhsingh) as a code owner April 10, 2026 13:04

github-actions bot added internal langchain For docs changes to LangChain oss labels Apr 10, 2026

Naomi Pentrel (npentrel) force-pushed the naomi/doc-925-change-langchain-quickstart branch from 45c4319 to 2421b07 Compare April 10, 2026 16:36

Mason Daugherty (mdrxy) reviewed Apr 12, 2026

View reviewed changes

Naomi Pentrel (npentrel) added 5 commits April 15, 2026 12:25

First draft

ce2d9d0

more models

917d447

swap example in python

71e7e9e

address feedback

98a4101

update js

3e82a6b

Naomi Pentrel (npentrel) force-pushed the naomi/doc-925-change-langchain-quickstart branch from 2421b07 to 3e82a6b Compare April 15, 2026 15:26

Naomi Pentrel (npentrel) commented Apr 16, 2026

View reviewed changes

Comment thread src/oss/langchain/quickstart.mdx Outdated

Naomi Pentrel (npentrel) force-pushed the naomi/doc-925-change-langchain-quickstart branch from 88b6856 to 4fe0b57 Compare April 16, 2026 17:48

github-actions bot added the python For content related to the Python version of LangChain projects label Apr 16, 2026

standardize

28e2ac3

Naomi Pentrel (npentrel) force-pushed the naomi/doc-925-change-langchain-quickstart branch from 4fe0b57 to 28e2ac3 Compare April 16, 2026 17:52

github-actions bot removed the python For content related to the Python version of LangChain projects label Apr 16, 2026

Tested and changed to other example

78bdc45

Naomi Pentrel (npentrel) requested a review from Mason Daugherty (mdrxy) April 17, 2026 09:40

Mason Daugherty (mdrxy) reviewed Apr 17, 2026

View reviewed changes

	- Deep agents: [Customization options](/oss/deepagents/customization), [persistent memory](/oss/deepagents/long-term-memory), [deploy to production](/oss/langgraph/deploy)
	- Deep Agents: [Customization options](/oss/deepagents/customization), [persistent memory](/oss/deepagents/long-term-memory), [deploy to production](/oss/langgraph/deploy)


		Now assemble your agent with all the components and run it.

		There are two different frameworks for creating agents: LangChain agents and deep agents.

Conversation

Naomi Pentrel (npentrel) commented Apr 10, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

github-actions bot commented Apr 10, 2026

Uh oh!

github-actions bot commented Apr 10, 2026

Uh oh!

github-actions bot commented Apr 10, 2026

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

github-actions bot commented Apr 15, 2026

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

github-actions bot commented Apr 16, 2026

Uh oh!

Lauren Hirata Singh (lnhsingh) commented Apr 16, 2026

Uh oh!

github-actions bot commented Apr 16, 2026

❌ Import check failed

Why this is a problem

How to fix

🤖 Automated check

Uh oh!

github-actions bot commented Apr 16, 2026

Uh oh!

github-actions bot commented Apr 16, 2026

Uh oh!

github-actions bot commented Apr 17, 2026

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

Naomi Pentrel (npentrel) commented Apr 10, 2026 •

edited

Loading