# Talking Shop **URL:** https://forum.langchain.com/c/talking-shop/12.md?page=4 [Latest](https://forum.langchain.com/latest.md) · [Categories](https://forum.langchain.com/categories.md) · [Tags](https://forum.langchain.com/tags.md) **Page:** 5 --- ## [zarq-langchain: Pre-trade crypto risk intelligence for LangChain agents](https://forum.langchain.com/t/zarq-langchain-pre-trade-crypto-risk-intelligence-for-langchain-agents/3098)
**Author:** [@Zarq](https://forum.langchain.com/u/Zarq)\ **Replies:** 0\ **Last updated:** [March 9, 2026, 1:38pm UTC](https://forum.langchain.com/t/zarq-langchain-pre-trade-crypto-risk-intelligence-for-langchain-agents/3098 "2026-03-09T13:38:09Z")
Just published zarq-langchain on PyPI — adds crypto risk scoring to any LangChain agent in 2 lines. pip install zarq-langchain from zarq\_langchain import ZARQRiskCheck tools = \[ZARQRiskCheck()\] What your agent gets p… --- ## [Merge or Die: a GitHub game for testing AI agents](https://forum.langchain.com/t/merge-or-die-a-github-game-for-testing-ai-agents/3089)
**Author:** [@aashmawy](https://forum.langchain.com/u/aashmawy)\ **Replies:** 0\ **Last updated:** [March 8, 2026, 1:14pm UTC](https://forum.langchain.com/t/merge-or-die-a-github-game-for-testing-ai-agents/3089 "2026-03-08T13:14:53Z")
I built something that I think is relevant for anyone working with agents, tools, and multi-step LLM workflows. It’s called Merge or Die: https://www.trajectly.dev/merge-or-die The idea is to make a real problem more … --- ## [Handling sensitive data in LangGraph's shared State (Found an interesting paper)](https://forum.langchain.com/t/handling-sensitive-data-in-langgraphs-shared-state-found-an-interesting-paper/3013)
**Author:** [@yagobski](https://forum.langchain.com/u/yagobski)\ **Replies:** 3\ **Last updated:** [March 1, 2026, 1:31pm UTC](https://forum.langchain.com/t/handling-sensitive-data-in-langgraphs-shared-state-found-an-interesting-paper/3013 "2026-03-01T13:31:16Z")
Hey everyone, I’ve been building some complex multi-agent workflows with LangGraph lately and hit a bit of a conceptual roadblock regarding data privacy: specifically, how to handle PII inside the shared State object. W… --- ## [Agent keeps thinking about calling non-existing get\_weather tool](https://forum.langchain.com/t/agent-keeps-thinking-about-calling-non-existing-get-weather-tool/2997)
**Author:** [@elhamin](https://forum.langchain.com/u/elhamin)\ **Replies:** 9\ **Last updated:** [February 26, 2026, 3:24am UTC](https://forum.langchain.com/t/agent-keeps-thinking-about-calling-non-existing-get-weather-tool/2997 "2026-02-26T03:24:14Z")
I’m running into a strange issue with my agents. I created an agent using createAgent and passed in a couple of tools. However, during reasoning, the agent keeps mentioning and attempting to call a non-existent tool call… --- ## [Source code vs. LangGraph first? ](https://forum.langchain.com/t/source-code-vs-langgraph-first/3009)
**Author:** [@yech](https://forum.langchain.com/u/yech)\ **Replies:** 5\ **Last updated:** [February 26, 2026, 1:55am UTC](https://forum.langchain.com/t/source-code-vs-langgraph-first/3009 "2026-02-26T01:55:13Z")
Many of my previous and even some current projects and new features were deeply developed using vLLM’s OpenAI Completion API or various model providers’ SDKs. I even wrote numerous strategy patterns to integrate differen… --- ## [Decentralized P2P coordination layer for LangChain agents — looking for feedback](https://forum.langchain.com/t/decentralized-p2p-coordination-layer-for-langchain-agents-looking-for-feedback/3017)
**Author:** [@Pointsnode](https://forum.langchain.com/u/Pointsnode)\ **Replies:** 1\ **Last updated:** [February 25, 2026, 5:05pm UTC](https://forum.langchain.com/t/decentralized-p2p-coordination-layer-for-langchain-agents-looking-for-feedback/3017 "2026-02-25T17:05:39Z")
Hey LangChain community :waving\_hand: I built HYPHA — an open source P2P infrastructure that lets AI agents discover each other via DHT, stream context over encrypted channels, and settle USDT payments through self-cust… --- ## [Chat model returns empty content when I inject a delayed ToolMessage from a scheduled callback](https://forum.langchain.com/t/chat-model-returns-empty-content-when-i-inject-a-delayed-toolmessage-from-a-scheduled-callback/1951)
**Author:** [@vicky70](https://forum.langchain.com/u/vicky70)\ **Replies:** 3\ **Last updated:** [February 20, 2026, 2:54am UTC](https://forum.langchain.com/t/chat-model-returns-empty-content-when-i-inject-a-delayed-toolmessage-from-a-scheduled-callback/1951 "2026-02-20T02:54:49Z")
Hey! I have a scheduler that triggers a tool function later (outside the original user–model exchange). When the scheduled time comes, it sends a ToolMessage back into the chat model — but the model returns an empty me… --- ## [Valkey Vector Store Support?](https://forum.langchain.com/t/valkey-vector-store-support/2966)
**Author:** [@jonathan.neufeld](https://forum.langchain.com/u/jonathan.neufeld)\ **Replies:** 5\ **Last updated:** [February 19, 2026, 9:02pm UTC](https://forum.langchain.com/t/valkey-vector-store-support/2966 "2026-02-19T21:02:17Z")
Has there been any interest or momentum on providing Valkey vector store support for LangChain in Python? Valkey is a high-performance open source alternative to Redis, backing ElastiCache, suitable for KV caching. Nor… --- ## [Groq Model Response](https://forum.langchain.com/t/groq-model-response/2820)
**Author:** [@razaullah](https://forum.langchain.com/u/razaullah)\ **Replies:** 9\ **Last updated:** [January 29, 2026, 9:52am UTC](https://forum.langchain.com/t/groq-model-response/2820 "2026-01-29T09:52:50Z")
I am using Groq as the model provider and the model I am using is openai/gpt-oss-120b. Some time I am getting a worse response from the model as follows. I am using a multi-agent flow. The response from the agent is: T… --- ## [Summarization Middleware](https://forum.langchain.com/t/summarization-middleware/2804)
**Author:** [@razaullah](https://forum.langchain.com/u/razaullah)\ **Replies:** 6\ **Last updated:** [January 26, 2026, 10:39am UTC](https://forum.langchain.com/t/summarization-middleware/2804 "2026-01-26T10:39:11Z")
I am posting with a new query. I have a multi-agent system flow using create\_agent as my supervisor and other sub\_agents exposed as a tools. The supervisor agent is defined as follows: supervisor = create\_agent( … --- ## [Why are the state variable active\_agent and the router node needed in the handoffs example?](https://forum.langchain.com/t/why-are-the-state-variable-active-agent-and-the-router-node-needed-in-the-handoffs-example/2772)
**Author:** [@guomo233](https://forum.langchain.com/u/guomo233)\ **Replies:** 4\ **Last updated:** [January 26, 2026, 7:54am UTC](https://forum.langchain.com/t/why-are-the-state-variable-active-agent-and-the-router-node-needed-in-the-handoffs-example/2772 "2026-01-26T07:54:51Z")
I was studying the handoffs example through the documentation \[ Handoffs - Docs by LangChain \] and noticed that when the agent returns a Command, it uses both gotoand the state variable active\_agent. The active\_agentis m… --- ## [Best ways to ensure sub‑agents follow long guides in a multi‑agent LangGraph system + questions about Todo List middleware](https://forum.langchain.com/t/best-ways-to-ensure-sub-agents-follow-long-guides-in-a-multi-agent-langgraph-system-questions-about-todo-list-middleware/2758)
**Author:** [@pigon224](https://forum.langchain.com/u/pigon224)\ **Replies:** 0\ **Last updated:** [January 20, 2026, 7:51am UTC](https://forum.langchain.com/t/best-ways-to-ensure-sub-agents-follow-long-guides-in-a-multi-agent-langgraph-system-questions-about-todo-list-middleware/2758 "2026-01-20T07:51:49Z")
Hi everyone, I’m building a complex multi-agent system, and I need each sub-agent to follow a detailed guide as closely as possible. The guides I’m using are long, ranging from 8,000 to 15,000 characters. I’m unsure abo… --- ## [Inter Communication between agents using create\_agent as a supervisor and as a tools](https://forum.langchain.com/t/inter-communication-between-agents-using-create-agent-as-a-supervisor-and-as-a-tools/2720)
**Author:** [@razaullah](https://forum.langchain.com/u/razaullah)\ **Replies:** 22\ **Last updated:** [January 18, 2026, 12:45pm UTC](https://forum.langchain.com/t/inter-communication-between-agents-using-create-agent-as-a-supervisor-and-as-a-tools/2720 "2026-01-18T12:45:59Z")
Context not shared across sub-agents despite same thread\_id Setup A supervisor agent created with create\_agent Multiple sub-agents created with create\_agent (patient\_agent, appt\_agent, book\_appointment) You expose each… --- ## [Multi-agent using create\_agent as a tools](https://forum.langchain.com/t/multi-agent-using-create-agent-as-a-tools/2686)
**Author:** [@razaullah](https://forum.langchain.com/u/razaullah)\ **Replies:** 3\ **Last updated:** [January 15, 2026, 4:08pm UTC](https://forum.langchain.com/t/multi-agent-using-create-agent-as-a-tools/2686 "2026-01-15T16:08:12Z")
I am working on a complex application, where I need to create a Customer AI Agent for HealthCare that will respond to user queries using the required tools. There multiple tasks for the agent, getting appointment details… --- ## [Is there a way to enable tool calling and structured output in the same agent?](https://forum.langchain.com/t/is-there-a-way-to-enable-tool-calling-and-structured-output-in-the-same-agent/2716)
**Author:** [@VBargas](https://forum.langchain.com/u/VBargas)\ **Replies:** 4\ **Last updated:** [January 13, 2026, 2:47pm UTC](https://forum.langchain.com/t/is-there-a-way-to-enable-tool-calling-and-structured-output-in-the-same-agent/2716 "2026-01-13T14:47:39Z")
I am having trouble using these two functionalities together (in the same agent). The goal is to allow the agent access to some tools, and then have the final response follow a defined Structured Output (OpenAI). When I … --- ## [Trouble Disabling Streamed Chunks](https://forum.langchain.com/t/trouble-disabling-streamed-chunks/2495)
**Author:** [@ethan-hemmaweg](https://forum.langchain.com/u/ethan-hemmaweg)\ **Replies:** 2\ **Last updated:** [December 15, 2025, 5:09pm UTC](https://forum.langchain.com/t/trouble-disabling-streamed-chunks/2495 "2025-12-15T17:09:21Z")
I can’t seem to disable the stream from agents. I’ve tried different combinations of streaming: false, disableStreaming: true, and tags: \[“nostream”\]. The final messages of this agent are not saved to the messages state,… --- ## [Summarization for Multi Agent Systems](https://forum.langchain.com/t/summarization-for-multi-agent-systems/2333)
**Author:** [@jriedel199715](https://forum.langchain.com/u/jriedel199715)\ **Replies:** 2\ **Last updated:** [November 26, 2025, 11:47am UTC](https://forum.langchain.com/t/summarization-for-multi-agent-systems/2333 "2025-11-26T11:47:14Z")
Are there any summarization tools/modules that come with langgraph when it comes to better managing context? I have a multi agent system where multiple agents can be tapped in to have conversations with a user. Over time… --- ## [Handoff goes back to initial agent](https://forum.langchain.com/t/handoff-goes-back-to-initial-agent/2238)
**Author:** [@reconlabs-sergio](https://forum.langchain.com/u/reconlabs-sergio)\ **Replies:** 3\ **Last updated:** [November 20, 2025, 9:12pm UTC](https://forum.langchain.com/t/handoff-goes-back-to-initial-agent/2238 "2025-11-20T21:12:22Z")
Hi, I’m trying to make the simplest possible example of a multiagent system with a handoff tool (as a surrogate of a more complex system). Agent A: addition and substraction Agent M: multiplication and division After… --- ## [My Deep Agents - E2B-CLI Tool](https://forum.langchain.com/t/my-deep-agents-e2b-cli-tool/2129)
**Author:** [@David-Barnes-Data-Im](https://forum.langchain.com/u/David-Barnes-Data-Im)\ **Replies:** 0\ **Last updated:** [November 7, 2025, 6:00pm UTC](https://forum.langchain.com/t/my-deep-agents-e2b-cli-tool/2129 "2025-11-07T18:00:41Z")
Hi all, So I noticed that every time I mention my Langgraph Deep Agents Psychology tool, I always get requests for the custom deep agents (the initial release) E2B mashup (which replaces the virtual filesystem with an E… --- ## [How to keep good architecture behind project with LLM and Python?](https://forum.langchain.com/t/how-to-keep-good-architecture-behind-project-with-llm-and-python/2127)
**Author:** [@aldruin](https://forum.langchain.com/u/aldruin)\ **Replies:** 0\ **Last updated:** [November 7, 2025, 11:17am UTC](https://forum.langchain.com/t/how-to-keep-good-architecture-behind-project-with-llm-and-python/2127 "2025-11-07T11:17:57Z")
Hello everyone. Today I want to show a question. As a Jr Developer, my main language was C#, and we think a lot about thinks like MVC, Clean Architecture, DDD… Recently I get my first job in the Development area, and … --- ## [Controlling Agent Behavior Based on Response Time Using LangChain Middleware](https://forum.langchain.com/t/controlling-agent-behavior-based-on-response-time-using-langchain-middleware/2123)
**Author:** [@Najiya](https://forum.langchain.com/u/Najiya)\ **Replies:** 0\ **Last updated:** [November 7, 2025, 4:09am UTC](https://forum.langchain.com/t/controlling-agent-behavior-based-on-response-time-using-langchain-middleware/2123 "2025-11-07T04:09:21Z")
Is it possible to control agent actions considering the time using Langchain Middleware?. I mean if certain response is taking more time, can i make the agent handle it the way i want using this concept? --- ## [Npm @langchain/anthropic version 0.3.3? What happen to 1.x?](https://forum.langchain.com/t/npm-langchain-anthropic-version-0-3-3-what-happen-to-1-x/2013)
**Author:** [@eljeffeg](https://forum.langchain.com/u/eljeffeg)\ **Replies:** 4\ **Last updated:** [November 1, 2025, 4:29pm UTC](https://forum.langchain.com/t/npm-langchain-anthropic-version-0-3-3-what-happen-to-1-x/2013 "2025-11-01T16:29:32Z")
Why did anthropic roll back the latest version? Shouldn’t it be going to 1.0.1? https://www.npmjs.com/package/@langchain/anthropic?activeTab=versions Package Current Wanted Latest @langcha… --- ## [Langchain create\_agent with tools parameter](https://forum.langchain.com/t/langchain-create-agent-with-tools-parameter/2030)
**Author:** [@mthekid](https://forum.langchain.com/u/mthekid)\ **Replies:** 1\ **Last updated:** [October 31, 2025, 10:28am UTC](https://forum.langchain.com/t/langchain-create-agent-with-tools-parameter/2030 "2025-10-31T10:28:03Z")
I’d like to welcome to LangChain version 1.0. I’m currently migrating my workflows to use create\_agent, but I noticed that when a tool specified in tools is called, the agent makes an additional LLM call after the norma… --- ## [Add WhatsApp Chat Document Loader (Python parity)](https://forum.langchain.com/t/add-whatsapp-chat-document-loader-python-parity/1884)
**Author:** [@Jude](https://forum.langchain.com/u/Jude)\ **Replies:** 3\ **Last updated:** [October 25, 2025, 8:32am UTC](https://forum.langchain.com/t/add-whatsapp-chat-document-loader-python-parity/1884 "2025-10-25T08:32:12Z")
Checked I searched existing ideas and did not find a similar one I added a very descriptive title I’ve clearly described the feature request and motivation for it Feature request Add a WhatsApp Chat Loader to LangCh… --- ## [thinkingBudget option for ChatGoogleGenerativeAI in langchainjs](https://forum.langchain.com/t/thinkingbudget-option-for-chatgooglegenerativeai-in-langchainjs/1909)
**Author:** [@harrygao56](https://forum.langchain.com/u/harrygao56)\ **Replies:** 0\ **Last updated:** [October 23, 2025, 7:48am UTC](https://forum.langchain.com/t/thinkingbudget-option-for-chatgooglegenerativeai-in-langchainjs/1909 "2025-10-23T07:48:57Z")
I’m using ChatGoogleGenerativeAI in the langchainjs SDK and need to be able to configure thinking budget. Seems like the python SDK has that option, as do the Vertex, Anthropic, and OpenAI models in the JS SDK. This is … --- ## [Strategies for Context Management](https://forum.langchain.com/t/strategies-for-context-management/1839)
**Author:** [@PeymanKh](https://forum.langchain.com/u/PeymanKh)\ **Replies:** 2\ **Last updated:** [October 16, 2025, 4:23pm UTC](https://forum.langchain.com/t/strategies-for-context-management/1839 "2025-10-16T16:23:34Z")
Hello friends, I am working on a multi-agent system and wanted to ask how you would approach a challenge I’m facing. I’m building a ReAct agent with a web search tool, and the topic it researches can require around 40 t… --- ## [LangGraph classification for agent discovery on HF Spaces - input needed](https://forum.langchain.com/t/langgraph-classification-for-agent-discovery-on-hf-spaces-input-needed/1757)
**Author:** [@ArchieCur](https://forum.langchain.com/u/ArchieCur)\ **Replies:** 0\ **Last updated:** [October 6, 2025, 12:38pm UTC](https://forum.langchain.com/t/langgraph-classification-for-agent-discovery-on-hf-spaces-input-needed/1757 "2025-10-06T12:38:48Z")
Hi LangGraph community, I’m building a discovery tool for AI agents and frameworks on HuggingFace Spaces, and I classified LangGraph as follows: A2-B2/B3-C2-D1-E2 • Multi-agent orchestration (A2) — though it supports … --- ## [Expecting more experience and best practice sharing about Claude Code practice](https://forum.langchain.com/t/expecting-more-experience-and-best-practice-sharing-about-claude-code-practice/1613)
**Author:** [@postbird](https://forum.langchain.com/u/postbird)\ **Replies:** 0\ **Last updated:** [September 22, 2025, 5:11pm UTC](https://forum.langchain.com/t/expecting-more-experience-and-best-practice-sharing-about-claude-code-practice/1613 "2025-09-22T17:11:26Z")
First, thanks to the langchain team for sharing your experience with claude code, I have read more that 5 times of this blog: https://blog.langchain.com/how-to-turn-claude-code-into-a-domain-specific-coding-agent/ And I… --- ## [Using functions or classes for node definition](https://forum.langchain.com/t/using-functions-or-classes-for-node-definition/1532)
**Author:** [@erjak](https://forum.langchain.com/u/erjak)\ **Replies:** 0\ **Last updated:** [September 15, 2025, 6:58am UTC](https://forum.langchain.com/t/using-functions-or-classes-for-node-definition/1532 "2025-09-15T06:58:36Z")
Hello! I’ve been using LangGraph for some time, but I’m unsure about the best practice: should nodes and agents be implemented with function definitions or with classes? --- ## [Feature request - Dart language text splitter](https://forum.langchain.com/t/feature-request-dart-language-text-splitter/1395)
**Author:** [@williambohrmann3](https://forum.langchain.com/u/williambohrmann3)\ **Replies:** 0\ **Last updated:** [August 28, 2025, 8:32pm UTC](https://forum.langchain.com/t/feature-request-dart-language-text-splitter/1395 "2025-08-28T20:32:53Z")
Hi. :waving\_hand: The Dart programming language, used for Flutter application, is quite popular. To my surprise, there is no splitter for Dart code Language — :parrot::link: LangChain documentation. I would like to be ab… [Previous page](https://forum.langchain.com/c/talking-shop/12.md?page=3) [Next page](https://forum.langchain.com/c/talking-shop/12.md?page=5)