--- title: "#crewai articles | Hashnode" description: "Explore 79 articles tagged #crewai on Hashnode. Tutorials, guides, and how-tos from the community." url: "https://hashnode.com/tag/crewai" --- Tag # #crewai 79 posts · 6 followers [MemorySync](https://hashnode.com/@memorysync) · [1d ago](https://hashnode.com/posts/crewai-memory-recall-cant-subtract-offset-naive-and-offset-aware-datetimes/6af341d6aa7c3d1e051b4c00) ### [CrewAI memory recall: "TypeError: can't subtract offset-naive and offset-aware datetimes"](https://memorysync.hashnode.dev/crewai-memory-recall-cant-subtract-offset-naive-and-offset-aware-datetimes?utm_source=hashnode&utm_medium=feed) CrewAI's recall scoring subtracts a naive utcnow() from your record's created\_at. A timezone-aware timestamp crashes recall. Repro and workaround. 6 min read · [memorysync.hashnode.dev](https://memorysync.hashnode.dev/crewai-memory-recall-cant-subtract-offset-naive-and-offset-aware-datetimes?utm_source=hashnode&utm_medium=feed) [MemorySync](https://hashnode.com/@memorysync) · [Nov 2](https://hashnode.com/posts/crewai-deep-recall-root-scope-leak/6ae8b728f76b02a8c20e7ea9) ### [CrewAI deep recall can return memories outside root\_scope: keep each crew's recall inside its own scope](https://memorysync.hashnode.dev/crewai-deep-recall-root-scope-leak?utm_source=hashnode&utm_medium=feed) In CrewAI 1.15.33, Memory.recall(depth="deep") searches every scope the LLM suggests, even outside the caller's root\_scope. Here's how to check, a settings-only mitigation, and a small storage wrapper that keeps reads inside one root. 14 min read · [memorysync.hashnode.dev](https://memorysync.hashnode.dev/crewai-deep-recall-root-scope-leak?utm_source=hashnode&utm_medium=feed) [MemorySync](https://hashnode.com/@memorysync) · [Nov 2](https://hashnode.com/posts/crewai-memory-scope-leading-slash-forget-reset/6ae8b6ed4c3de7cb65a710a6) ### [CrewAI memory: forget() and reset() miss records whose scope has no leading slash](https://memorysync.hashnode.dev/crewai-memory-scope-leading-slash-forget-reset?utm_source=hashnode&utm_medium=feed) In CrewAI 1.15.x, a scope written without a leading slash is stored as-is, and recall, forget and reset don't agree on which records it names. Anchor every scope yourself until the fix lands. 12 min read · [memorysync.hashnode.dev](https://memorysync.hashnode.dev/crewai-memory-scope-leading-slash-forget-reset?utm_source=hashnode&utm_medium=feed) [MemorySync](https://hashnode.com/@memorysync) · [Oct 30](https://hashnode.com/posts/crewai-memory-true-please-provide-an-openai-api-key/6ae4c349a0d8cb89ef9d0b2c) ### [CrewAI memory=True: "Please provide an OpenAI API key" when you're not using OpenAI (embedder fix)](https://memorysync.hashnode.dev/crewai-memory-true-please-provide-an-openai-api-key?utm_source=hashnode&utm_medium=feed) Fix CrewAI's 'Please provide an OpenAI API key' error after memory=True. Memory embeds with OpenAI by default; set an embedder like Ollama instead. 6 min read · [memorysync.hashnode.dev](https://memorysync.hashnode.dev/crewai-memory-true-please-provide-an-openai-api-key?utm_source=hashnode&utm_medium=feed) [MemorySync](https://hashnode.com/@memorysync) · [Oct 23](https://hashnode.com/posts/crewai-persistent-memory-memorysync/6adb8719f6f60b8802f03fb6) ### [Persistent memory for CrewAI 1.x: keep crew memory across runs and redeploys](https://memorysync.hashnode.dev/crewai-persistent-memory-memorysync?utm_source=hashnode&utm_medium=feed) Disclosure: I'm Mohammed, founder of MemorySync. I drafted this article with AI assistance and reviewed it for accuracy against the official MemorySync and CrewAI docs. The code is adapted from the do 7 min read · [memorysync.hashnode.dev](https://memorysync.hashnode.dev/crewai-persistent-memory-memorysync?utm_source=hashnode&utm_medium=feed) [AgentSearch](https://hashnode.com/@agentsearchhq) · [Oct 19](https://hashnode.com/posts/add-web-search-to-crewai-agent/6ad62bb45902946fa4cf9757) ### [How to Add Web Search to a CrewAI Agent (Free, No API Key)](https://agentsearchhq.hashnode.dev/add-web-search-to-crewai-agent?utm_source=hashnode&utm_medium=feed) Add web search to a CrewAI agent with crewai-agentsearch: free search with no API key, a researcher + writer crew, Extract, Render and spend caps. 13 min read · [agentsearchhq.hashnode.dev](https://agentsearchhq.hashnode.dev/add-web-search-to-crewai-agent?utm_source=hashnode&utm_medium=feed) [rambo](https://hashnode.com/@rambo-zambo) · [Sep 25](https://hashnode.com/posts/give-your-langchain-and-crewai-agents-verifiable-execution-receipts/6ab67de9bcf6998715c2139d) ### [Give Your LangChain and CrewAI Agents Verifiable Execution Receipts](https://rambo-zambo.hashnode.dev/give-your-langchain-and-crewai-agents-verifiable-execution-receipts?utm_source=hashnode&utm_medium=feed) Two small PyPI packages hook LangChain callbacks and CrewAI task listeners to mint verifiable execution receipts. Free tier, no account. 9 min read · [rambo-zambo.hashnode.dev](https://rambo-zambo.hashnode.dev/give-your-langchain-and-crewai-agents-verifiable-execution-receipts?utm_source=hashnode&utm_medium=feed) [Praise James](https://hashnode.com/@thepraisejames) · [Aug 31](https://hashnode.com/posts/replace-crewai-memory-in-production-with-vectorai-db/6a953acb89c044d1876c80fd) ### [Replace CrewAI Memory in Production with VectorAI DB](https://actiandev.hashnode.dev/replace-crewai-memory-in-production-with-vectorai-db?utm_source=hashnode&utm_medium=feed) If you've deployed a CrewAI application with memory=True and you're seeing "database is locked" errors under concurrent load, lost memory after a container restart, or memory bleeding between users in 13 min read · [actiandev.hashnode.dev](https://actiandev.hashnode.dev/replace-crewai-memory-in-production-with-vectorai-db?utm_source=hashnode&utm_medium=feed) [Ratish Jain](https://hashnode.com/@Ratishjain) · [Aug 16](https://hashnode.com/posts/deploying-a-crewai-multi-agent-crew-to-amazon-bedrock-agentcore/6a81a974dc62586aa047d0d5) ### [Deploying a CrewAI Multi-Agent Crew to Amazon Bedrock AgentCore](https://blog.ratishfolio.com/deploying-a-crewai-multi-agent-crew-to-amazon-bedrock-agentcore?utm_source=hashnode&utm_medium=feed) AWS AgentCore Guide 8 min read · [blog.ratishfolio.com](https://blog.ratishfolio.com/deploying-a-crewai-multi-agent-crew-to-amazon-bedrock-agentcore?utm_source=hashnode&utm_medium=feed) [Debashish Ghosal](https://hashnode.com/@deghosal) · [Aug 13](https://hashnode.com/posts/contextual-risk-for-agent-tool-calls-a-4-state-permission-engine-field-tested-across-10-frameworks/6a7d6ad1ade5b2226985db01) ### [Contextual Risk for Agent Tool Calls: A 4-State Permission Engine Field-Tested Across 10 Frameworks](https://pragmatic-engineer.hashnode.dev/contextual-risk-for-agent-tool-calls-a-4-state-permission-engine-field-tested-across-10-frameworks?utm_source=hashnode&utm_medium=feed) A permission engine that returns allow, audit, escalate, or deny based on runtime context, field-tested across 10 agent frameworks. 12 min read · [pragmatic-engineer.hashnode.dev](https://pragmatic-engineer.hashnode.dev/contextual-risk-for-agent-tool-calls-a-4-state-permission-engine-field-tested-across-10-frameworks?utm_source=hashnode&utm_medium=feed) [2](https://hashnode.com/posts/contextual-risk-for-agent-tool-calls-a-4-state-permission-engine-field-tested-across-10-frameworks/6a7d6ad1ade5b2226985db01)