Add Langchain quickstart guide and notebook demo #95
Conversation
WalkthroughAdds two new educational resources: a markdown guide explaining how to create an AI agent with Langchain, covering prerequisites, quickstart steps, tool definition, model initialization, agent creation, and deployment; plus a corresponding Jupyter notebook demonstrating end-to-end setup with a weather tool, FastAPI integration, and async agent execution. Changes
Estimated code review effort🎯 2 (Simple) | ⏱️ ~10 minutes Possibly related PRs
Suggested reviewers
Poem
Pre-merge checks✅ Passed checks (3 passed)
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. Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 3
📜 Review details
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (1)
docs/en/solutions/How_to_Create_an_AI_Agent_with_Langchain.md
🧰 Additional context used
🪛 GitHub Actions: Build and Update
docs/en/solutions/How_to_Create_an_AI_Agent_with_Langchain.md
[error] 1-1: MDX compile error: Language ipypb is not included in this bundle. You may want to load it from external source.
🪛 markdownlint-cli2 (0.18.1)
docs/en/solutions/How_to_Create_an_AI_Agent_with_Langchain.md
10-10: No space after hash on atx style heading
(MD018, no-missing-space-atx)
davidwtf
force-pushed
the
feat/langchain_quick_start
branch
from
414b3c8 to
f0d1460
Compare
There was a problem hiding this comment.
Actionable comments posted: 1
🧹 Nitpick comments (3)
docs/public/langchain/langchain_quickstart.ipynb (3)
262-270: Remove duplicate import ofrequests.
requestsis already imported in the earlier imports cell (line 52). This redefinition is unnecessary.🔎 Suggested fix
# Import FastAPI components from fastapi import FastAPI from pydantic import BaseModel import uvicorn from threading import Thread import time -import requests
351-361: Add timeout torequests.postcall.The HTTP request lacks a timeout, which could cause the notebook cell to hang indefinitely if the server is unresponsive.
Also, line 359 has an f-string without placeholders that can be simplified.
🔎 Suggested fix
# Test the API endpoint response = requests.post( "http://127.0.0.1:8000/chat", - json={"message": "What's the weather in Shanghai?"} + json={"message": "What's the weather in Shanghai?"}, + timeout=30 ) print(f"Status Code: {response.status_code}") -print(f"Response:") +print("Response:") print(response.json().get('response'))
314-315: Remove duplicateuvicornimport.
uvicornis already imported in the previous cell. Consider consolidating imports.🔎 Suggested fix
# Start server in background thread (for notebook demonstration) # In production, run: uvicorn langchain_demo.api:app --host 0.0.0.0 --port 8000 -import uvicorn from uvicorn import Config, Server
📜 Review details
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (2)
docs/en/solutions/How_to_Create_an_AI_Agent_with_Langchain.mddocs/public/langchain/langchain_quickstart.ipynb
✅ Files skipped from review due to trivial changes (1)
- docs/en/solutions/How_to_Create_an_AI_Agent_with_Langchain.md
🧰 Additional context used
🪛 Ruff (0.14.10)
docs/public/langchain/langchain_quickstart.ipynb
51-51: Do not catch blind exception: Exception
(BLE001)
52-52: Use explicit conversion flag
Replace with conversion flag
(RUF010)
77-81: await statement outside of a function
(F704)
103-107: await statement outside of a function
(F704)
119-119: Redefinition of unused requests from line 13
Remove definition: requests
(F811)
152-152: Redefinition of unused uvicorn from line 116
Remove definition: uvicorn
(F811)
179-179: Probable use of requests call without timeout
(S113)
185-185: f-string without any placeholders
Remove extraneous f prefix
(F541)
🔇 Additional comments (3)
docs/public/langchain/langchain_quickstart.ipynb (3)
72-79: LGTM!The configuration pattern with environment variable fallbacks is appropriate for a tutorial notebook.
95-122: LGTM!The weather tool implementation is well-structured for a tutorial:
- Includes proper timeout on the HTTP request (30s)
- Returns structured error information on failure
- The broad exception catch is acceptable for demo purposes to ensure graceful degradation
48-48: No action required. Thecreate_agentfunction is a valid and documented export fromlangchain.agents. It constructs a production-ready agent graph and is the standard general-purpose agent creation function in LangChain. The import will succeed without error.
There was a problem hiding this comment.
Actionable comments posted: 2
♻️ Duplicate comments (1)
docs/public/langchain/langchain_quickstart.ipynb (1)
331-331: Missing/healthendpoint referenced in output.The print statement references
http://127.0.0.1:8000/health, but no/healthendpoint is defined. This will cause confusion when users try to access it.
🧹 Nitpick comments (1)
docs/public/langchain/langchain_quickstart.ipynb (1)
268-268: Duplicate import ofrequests.
requestsis already imported in cell 2 (line 52). This redefinition is unnecessary.🔎 Remove duplicate import
# Import FastAPI components from fastapi import FastAPI from pydantic import BaseModel from threading import Thread import time -import requests
📜 Review details
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (1)
docs/public/langchain/langchain_quickstart.ipynb
🧰 Additional context used
🪛 Ruff (0.14.10)
docs/public/langchain/langchain_quickstart.ipynb
51-51: Do not catch blind exception: Exception
(BLE001)
52-52: Use explicit conversion flag
Replace with conversion flag
(RUF010)
77-81: await statement outside of a function
(F704)
103-107: await statement outside of a function
(F704)
117-117: Redefinition of unused requests from line 13
Remove definition: requests
(F811)
175-175: Probable use of requests call without timeout
(S113)
181-181: f-string without any placeholders
Remove extraneous f prefix
(F541)
🔇 Additional comments (1)
docs/public/langchain/langchain_quickstart.ipynb (1)
48-48: No issue. Thecreate_agentfunction fromlangchain.agentsis valid in LangChain v1.0+, which matches the notebook's dependency specification (langchain>=1.0.0). This is the official entrypoint for agent creation in current LangChain versions.
There was a problem hiding this comment.
Actionable comments posted: 0
🧹 Nitpick comments (3)
theme/layout/index.tsx (3)
74-93: Consider user feedback for download failures.The same-origin download logic is correct, but failures are only logged to the console with no user-visible feedback. Depending on UX requirements, you may want to add a toast notification or error message.
Additionally, the 100ms cleanup timeout (line 91) is a magic number that could be extracted as a named constant for maintainability.
105-121: Improve type safety for event handler.The click event handler uses a generic
Eventtype (line 114), but it should be typed asMouseEventfor better type safety and to access mouse-specific properties if needed in the future.🔎 Suggested type improvement
- link.addEventListener('click', (e: Event) => { + link.addEventListener('click', (e: MouseEvent) => { e.preventDefault(); e.stopPropagation(); e.stopImmediatePropagation(); downloadFile(href, pathname.split('/').pop() || 'download'); }, { capture: true });
57-130: Optional: Extract magic number for timeout delays.The value
100(milliseconds) appears multiple times as a timeout delay (lines 91, 101, 124). Consider extracting it as a named constant for better maintainability and consistency.🔎 Suggested constant extraction
// Handle file download links useEffect(() => { const DOWNLOAD_EXTENSIONS = ['.ipynb', '.zip', '.tgz', '.tar.gz', '.sh', '.py', '.sql']; + const CLEANUP_DELAY_MS = 100; // ... rest of code ... setTimeout(() => { window.URL.revokeObjectURL(blobUrl); document.body.removeChild(a); - }, 100); + }, CLEANUP_DELAY_MS); }) .catch(() => console.warn('Failed to download file:', url)); } else { const a = document.createElement('a'); a.href = url; a.download = filename; a.style.display = 'none'; document.body.appendChild(a); a.click(); - setTimeout(() => document.body.removeChild(a), 100); + setTimeout(() => document.body.removeChild(a), CLEANUP_DELAY_MS); } }; handleDownloadLinks(); - const observer = new MutationObserver(() => setTimeout(handleDownloadLinks, 100)); + const observer = new MutationObserver(() => setTimeout(handleDownloadLinks, CLEANUP_DELAY_MS));
📜 Review details
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (1)
theme/layout/index.tsx
⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (1)
- GitHub Check: build
🔇 Additional comments (3)
theme/layout/index.tsx (3)
57-72: LGTM!The extension list and helper functions are well-structured. The fallback pathname parsing is reasonable for edge cases where URL construction fails.
94-102: Verify cross-origin download behavior.The
downloadattribute on cross-origin anchors may not trigger downloads as expected due to browser CORS restrictions. Without proper CORS headers (Content-Disposition,Access-Control-Allow-Origin), browsers typically open the file in a new tab instead of downloading it.Consider testing this behavior with your cross-origin resources, or add a fallback mechanism if downloads fail.
123-130: Potential race condition with debounced MutationObserver.The MutationObserver callback is debounced with a 100ms delay (line 124), which theoretically could allow a user to click a newly added download link before the handler is attached. However, this is unlikely in practice since content mutations typically occur during page navigation or initial load, not during active user interaction.
If you observe issues with links not triggering downloads, consider removing the
setTimeoutdelay or reducing it significantly.
davidwtf
force-pushed
the
feat/langchain_quick_start
branch
from
b00fbb3 to
b6aa8ec
Compare
There was a problem hiding this comment.
Actionable comments posted: 0
♻️ Duplicate comments (3)
docs/public/langchain/langchain_quickstart.ipynb (3)
48-48:create_agentimport likely fails - use standard LangChain agent pattern.This import references a non-standard function. LangChain typically uses
create_react_agent,create_openai_functions_agent, or similar specific creators, not a genericcreate_agent.Verify the correct LangChain agent creation API:
What is the correct way to create an agent in LangChain 1.0+ with tools and a system prompt? Does create_agent exist in langchain.agents?
164-171: Agent creation uses non-standard API.The
create_agentfunction with parametersmodel,tools, andsystem_promptdoesn't match standard LangChain patterns. Typical usage requires:
- A prompt template (from
hub.pull()orChatPromptTemplate)create_react_agent(llm, tools, prompt)or similar- Wrapping in
AgentExecutor(agent=agent, tools=tools)🔎 Standard LangChain pattern
-agent = create_agent( - model=llm, - tools=[get_weather], - system_prompt="You are a helpful weather assistant." -) +from langchain.agents import create_react_agent, AgentExecutor +from langchain_core.prompts import ChatPromptTemplate + +prompt = ChatPromptTemplate.from_messages([ + ("system", "You are a helpful weather assistant."), + ("human", "{input}"), + ("placeholder", "{agent_scratchpad}"), +]) + +agent = create_react_agent(llm, [get_weather], prompt) +agent_executor = AgentExecutor(agent=agent, tools=[get_weather], verbose=True)Note: You'll also need to update the invocation calls to use
agent_executorinstead ofagent.
348-351: Add timeout to prevent indefinite hang.The
requests.postcall lacks a timeout parameter, which can cause the cell to hang if the server is unresponsive.🔎 Proposed fix
response = requests.post( "http://127.0.0.1:8000/chat", - json={"message": "What's the weather in Shanghai?"} + json={"message": "What's the weather in Shanghai?"}, + timeout=10 )
🧹 Nitpick comments (2)
docs/public/langchain/langchain_quickstart.ipynb (2)
268-268: Remove redundantrequestsimport.The
requestsmodule is already imported at line 52. This redefinition is unnecessary.🔎 Proposed fix
from fastapi import FastAPI from pydantic import BaseModel from threading import Thread import time -import requests
354-354: Remove unnecessary f-string prefix.The string has no placeholders, so the
fprefix is not needed.🔎 Proposed fix
print(f"Status Code: {response.status_code}") -print(f"Response:") +print("Response:") print(response.json().get('response'))
📜 Review details
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (3)
docs/en/solutions/How_to_Create_an_AI_Agent_with_Langchain.mddocs/public/langchain/langchain_quickstart.ipynbtheme/layout/index.tsx
🚧 Files skipped from review as they are similar to previous changes (1)
- theme/layout/index.tsx
🧰 Additional context used
🪛 Ruff (0.14.10)
docs/public/langchain/langchain_quickstart.ipynb
51-51: Do not catch blind exception: Exception
(BLE001)
52-52: Use explicit conversion flag
Replace with conversion flag
(RUF010)
77-81: await statement outside of a function
(F704)
103-107: await statement outside of a function
(F704)
117-117: Redefinition of unused requests from line 13
Remove definition: requests
(F811)
173-173: Probable use of requests call without timeout
(S113)
179-179: f-string without any placeholders
Remove extraneous f prefix
(F541)
🔇 Additional comments (1)
docs/en/solutions/How_to_Create_an_AI_Agent_with_Langchain.md (1)
1-38: LGTM! Well-structured documentation.The documentation is clear, well-organized, and provides helpful context for users. The front matter is properly formatted, and the content effectively guides users to the quickstart notebook and additional resources.
davidwtf
changed the title
There was a problem hiding this comment.
Actionable comments posted: 0
🧹 Nitpick comments (1)
docs/public/langchain/langchain_quickstart.ipynb (1)
268-268: Optional: Remove duplicate import.The
requestsmodule is already imported at line 52. This reimport is unnecessary.🔎 Proposed fix
# Import FastAPI components from fastapi import FastAPI from pydantic import BaseModel from threading import Thread import time -import requests
📜 Review details
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (1)
docs/public/langchain/langchain_quickstart.ipynb
🧰 Additional context used
🧠 Learnings (2)
📓 Common learnings
Learnt from: davidwtf
Repo: alauda/knowledge PR: 95
File: docs/public/langchain/langchain_quickstart.ipynb:163-171
Timestamp: 2025-12-30T08:43:22.275Z
Learning: In LangChain, `create_agent(model=..., tools=[...], system_prompt="...")` is a valid agent creation pattern where `model` accepts a chat model instance (e.g., ChatOpenAI) and `system_prompt` accepts a string for system instructions.
📚 Learning: 2025-12-30T08:43:22.275Z
Learnt from: davidwtf
Repo: alauda/knowledge PR: 95
File: docs/public/langchain/langchain_quickstart.ipynb:163-171
Timestamp: 2025-12-30T08:43:22.275Z
Learning: When reviewing LangChain notebooks, validate the agent creation pattern: ensure code uses create_agent(model=..., tools=[...], system_prompt="...") with model being a chat model instance (e.g., ChatOpenAI) and system_prompt provided as a string describing system instructions. This reflects a standard agent construction and can be applied across LangChain notebook files.
Applied to files:
docs/public/langchain/langchain_quickstart.ipynb
🪛 Ruff (0.14.10)
docs/public/langchain/langchain_quickstart.ipynb
51-51: Do not catch blind exception: Exception
(BLE001)
52-52: Use explicit conversion flag
Replace with conversion flag
(RUF010)
77-81: await statement outside of a function
(F704)
103-107: await statement outside of a function
(F704)
117-117: Redefinition of unused requests from line 13
Remove definition: requests
(F811)
🔇 Additional comments (13)
docs/public/langchain/langchain_quickstart.ipynb (13)
25-25: LGTM - Dependencies and installation approach.The pip install command with version constraints and custom target directory is appropriate for a notebook demonstration environment.
41-55: LGTM - Path setup and imports.The sys.path manipulation correctly supports the custom package installation directory, and all imports are appropriate for the notebook's functionality.
73-78: LGTM - Configuration setup.The environment variable configuration with sensible defaults is appropriate for a tutorial notebook.
96-121: LGTM - Weather tool implementation.The tool is well-structured with proper error handling, timeout, and documentation. The broad exception catch at line 118 is appropriate for a demo tool as it ensures consistent error responses.
139-146: LGTM - LLM initialization.The ChatOpenAI configuration is correct with appropriate parameters for a deterministic demo.
164-170: LGTM - Agent creation.The agent creation pattern is correct. Based on learnings,
create_agentwithmodel,tools, andsystem_promptparameters is valid LangChain API usage.
188-194: LGTM - Agent invocation.The async invocation pattern is correct. Top-level
awaitis valid in Jupyter notebooks with async kernel support.
210-220: LGTM - Result display helper.The defensive programming with type and attribute checks ensures robust handling of various result formats.
238-245: LGTM - Second query demonstration.The invocation follows the same correct pattern as the first query.
276-289: LGTM - Chat endpoint implementation.The endpoint correctly invokes the agent and handles response formatting with proper fallback logic.
310-329: LGTM - Server startup in notebook context.The daemon thread approach with startup delay is appropriate for notebook demonstrations, with clear documentation about production considerations.
347-356: LGTM - API test with proper timeout.The HTTP request includes an appropriate timeout parameter, addressing reliability concerns.
374-380: LGTM - Graceful server shutdown.The defensive checks and clear user feedback provide a robust shutdown mechanism.
theme/layout/index.tsx
Outdated
| }, []); | ||
|
|
||
| // Handle file download links | ||
| useEffect(() => { |
There was a problem hiding this comment.
整体逻辑没啥问题，通过直接查询文档的方式重载 a 标签的方式不太好， rspress 有提供通过插槽自定义组件的方式来重载 MDX 中的 a 标签渲染
用法是
import {
Badge,
LastUpdated,
Layout,
getCustomMDXComponent,
} from "@rspress/core/theme";
<Layout
uiSwitch={uiSwitch}
HomeLayout={HomeLayout}
beforeDocContent={
<>
<BreadCrumb></BreadCrumb>
</>
}
beforeDocFooter={<Badges></Badges>}
afterDocFooter={
<div className="flex justify-between">
<LastUpdated></LastUpdated>
<DocID></DocID>
</div>
}
beforeOutline={<EditOnGithub></EditOnGithub>}
components={{
a: (props: any) => {
const CustomMDXComponent = getCustomMDXComponent();
const pathname = getPathname(props.href);
if (!shouldDownload(pathname)) {
return <CustomMDXComponent.a {...props}></CustomMDXComponent.a>;
}
return (
<CustomMDXComponent.a
{...props}
href=""
onClick={(e: Event) => {
e.preventDefault();
e.stopPropagation();
e.stopImmediatePropagation?.();
downloadFile(
props.href,
pathname.split("/").pop() || "download"
);
}}
></CustomMDXComponent.a>
);
},
}}
></Layout>
| "metadata": {}, | ||
| "outputs": [], | ||
| "source": [ | ||
| "!pip install \"langchain>=1.0.0\" \"langchain-openai>=1.0.0\" \"requests\" \"fastapi\" \"uvicorn\" --target ~/packages" |
There was a problem hiding this comment.
好像 langchain > 1.0 需要 python > 3.13 ?，现有的 workbench 的 python 是 3.11 需要升级notebook镜像么
There was a problem hiding this comment.
不用，用开发环境的notebook测过了，可以跑。
不过比较麻烦的是，notebook中自带了0.x的langchain，直接pip安装新版本的位置不是python加载的位置。
为解决这个问题，pip install 指定了安装目录，import前修改了sys.path。
感觉 notebook 镜像后面可以优化一下。
| "source": [ | ||
| "## 6. Create Agent\n", | ||
| "\n", | ||
| "Create an agent with the weather tool:\n" |
There was a problem hiding this comment.
增加个跳转 lanchain 文档的链接？比如实现更复杂的场景参考...
There was a problem hiding this comment.
文章中写了langchain的文档连接了。这里也可以再加一下，我加下吧。
davidwtf
force-pushed
the
feat/langchain_quick_start
branch
from
97c7253 to
83ea149
Compare
davidwtf
changed the title
There was a problem hiding this comment.
Actionable comments posted: 0
🧹 Nitpick comments (1)
docs/public/langchain/langchain_quickstart.ipynb (1)
268-268: Optional cleanup: Redundant import.The
requestsmodule is already imported on line 52. While re-importing in notebook cells is harmless and sometimes done for cell independence, this particular import is unnecessary since earlier cells already loaded it.🔎 Optional: Remove redundant import
# Import FastAPI components from fastapi import FastAPI from pydantic import BaseModel from threading import Thread import time -import requests
📜 Review details
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (2)
docs/en/solutions/How_to_Create_an_AI_Agent_with_Langchain.mddocs/public/langchain/langchain_quickstart.ipynb
🚧 Files skipped from review as they are similar to previous changes (1)
- docs/en/solutions/How_to_Create_an_AI_Agent_with_Langchain.md
🧰 Additional context used
🧠 Learnings (2)
📓 Common learnings
Learnt from: davidwtf
Repo: alauda/knowledge PR: 95
File: docs/public/langchain/langchain_quickstart.ipynb:163-171
Timestamp: 2025-12-30T08:43:32.084Z
Learning: In LangChain, `create_agent(model=..., tools=[...], system_prompt="...")` is a valid agent creation pattern where `model` accepts a chat model instance (e.g., ChatOpenAI) and `system_prompt` accepts a string for system instructions.
📚 Learning: 2025-12-30T08:43:22.275Z
Learnt from: davidwtf
Repo: alauda/knowledge PR: 95
File: docs/public/langchain/langchain_quickstart.ipynb:163-171
Timestamp: 2025-12-30T08:43:22.275Z
Learning: When reviewing LangChain notebooks, validate the agent creation pattern: ensure code uses create_agent(model=..., tools=[...], system_prompt="...") with model being a chat model instance (e.g., ChatOpenAI) and system_prompt provided as a string describing system instructions. This reflects a standard agent construction and can be applied across LangChain notebook files.
Applied to files:
docs/public/langchain/langchain_quickstart.ipynb
🪛 Ruff (0.14.10)
docs/public/langchain/langchain_quickstart.ipynb
51-51: Do not catch blind exception: Exception
(BLE001)
52-52: Use explicit conversion flag
Replace with conversion flag
(RUF010)
77-81: await statement outside of a function
(F704)
103-107: await statement outside of a function
(F704)
117-117: Redefinition of unused requests from line 13
Remove definition: requests
(F811)
🔇 Additional comments (8)
docs/public/langchain/langchain_quickstart.ipynb (8)
41-56: LGTM! Clean import setup with custom package path.The sys.path manipulation to use the custom package location is well-implemented and handles the notebook environment constraints appropriately. The imports are complete and correctly organized.
73-78: LGTM! Environment configuration is clear and flexible.The configuration correctly uses environment variables with appropriate defaults. The placeholder text for the API key makes it clear what users need to provide.
96-122: LGTM! Well-implemented weather tool with proper error handling.The tool function correctly:
- Uses the
@tooldecorator for LangChain integration- Includes timeout on the HTTP request (line 108)
- Returns structured dictionaries for both success and error cases
- Provides clear docstring with Args and Returns
The broad
Exceptioncatch (line 118) flagged by static analysis is appropriate here—for a demo tool, gracefully handling all errors and returning a structured error response is preferable to raising exceptions.
139-146: LGTM! LLM initialization is correct.The ChatOpenAI initialization properly handles the optional base_url with a truthiness check, allowing flexibility between OpenAI and alternative providers.
164-170: LGTM! Agent creation follows correct LangChain pattern.The agent creation correctly uses
create_agentwith themodel,tools, andsystem_promptparameters as confirmed by LangChain documentation and previous testing.Based on learnings: The
create_agent(model=..., tools=[...], system_prompt="...")pattern is validated.
188-194: LGTM! Agent execution and result display work correctly.The async agent invocation using
awaitin notebook cells is fully supported by IPython/Jupyter environments (the static analysis warnings about "await outside function" are false positives for notebooks).The
print_resulthelper function correctly handles the agent response structure by checking for messages and extracting the last message content, with appropriate fallbacks.Also applies to: 210-220, 239-245
263-292: LGTM! FastAPI service implementation is well-structured.The FastAPI integration correctly:
- Defines a proper Pydantic model for request validation (lines 273-274)
- Implements async endpoint that properly invokes the agent (lines 276-289)
- Uses daemon thread for notebook demonstration with clear documentation (lines 310-329)
- Includes timeout on the test request (line 351) as addressed from previous review
- Provides graceful server shutdown (lines 374-380)
The implementation appropriately balances demonstration purposes with production-ready patterns, and includes helpful comments explaining the threading approach is for notebook use.
Also applies to: 310-330, 347-356, 374-380
394-400: LGTM! Helpful resources section.The resources section provides valuable links to official LangChain documentation and educational materials, making it easier for users to explore more advanced scenarios. This nicely addresses previous feedback about providing documentation references.
Summary by CodeRabbit
Release Notes
✏️ Tip: You can customize this high-level summary in your review settings.