CLAUDE LABJP
FORK — Claude Code 2.1.212 changes what /fork does: it copies your conversation into a new background session with its own row in claude agents, so you can keep working. The old in-session subagent is now /subtaskLIMITS — WebSearch calls are now capped at 200 per session by default, and subagent spawns get the same 200 ceiling, so a runaway search or delegation loop stops on its ownMCPBG — MCP tool calls running past two minutes now move to the background automatically, keeping the session usable. Tune the threshold with CLAUDE_CODE_MCP_AUTO_BACKGROUND_MSPLANFIX — Fixed plan mode auto-running file-modifying Bash commands such as touch and rm without a permission prompt or an SDK canUseTool callbackSONNET5 — Claude Sonnet 5 is running on introductory pricing of $2 per million input tokens and $10 per million output. After August 31 it moves to $3 and $15IPO — Bankers are reportedly lining up investor meetings for Anthropic ahead of a possible public listing as soon as OctoberFORK — Claude Code 2.1.212 changes what /fork does: it copies your conversation into a new background session with its own row in claude agents, so you can keep working. The old in-session subagent is now /subtaskLIMITS — WebSearch calls are now capped at 200 per session by default, and subagent spawns get the same 200 ceiling, so a runaway search or delegation loop stops on its ownMCPBG — MCP tool calls running past two minutes now move to the background automatically, keeping the session usable. Tune the threshold with CLAUDE_CODE_MCP_AUTO_BACKGROUND_MSPLANFIX — Fixed plan mode auto-running file-modifying Bash commands such as touch and rm without a permission prompt or an SDK canUseTool callbackSONNET5 — Claude Sonnet 5 is running on introductory pricing of $2 per million input tokens and $10 per million output. After August 31 it moves to $3 and $15IPO — Bankers are reportedly lining up investor meetings for Anthropic ahead of a possible public listing as soon as October
Articles/Claude.ai
Claude.ai/2026-03-18Beginner

Claude Memory Now Available to All Users — AI That Learns from Your Chat History

Complete guide to Claude's new Memory feature. Learn how Claude automatically remembers your preferences from chat history, now available to all users including the free plan. Setup, privacy controls, and practical tips.

Claude46Memory3Chat HistoryPersonalization2Free Plan2026 Latest

import { Callout } from '@/components/ui/callout';

March 2026 brought a significant milestone for Claude users everywhere. Anthropic announced the availability of Memory — a powerful feature that allows Claude to learn and remember your preferences from chat history — to all users, including those on the free plan.

Previously, this capability was restricted to paid users. Now, anyone using Claude can benefit from a more personalized, context-aware AI assistant. This comprehensive guide covers everything you need to know about Claude's Memory feature.

What is Claude's Memory Feature?

At its core, Claude's Memory is elegantly simple: Claude learns your preferences, work style, and interests from your chat history and remembers them across conversations.

Traditionally, starting a new chat meant explaining the same context repeatedly. "I'm a data scientist working with Python," "I prefer concise answers," "I'm building a mobile app." With Memory, you tell Claude once, and it remembers.

ℹ️
**Real-world example:** Tell Claude "I'm a frontend developer specializing in Vue.js and Tailwind CSS" in one conversation. In your next conversation three weeks later, when you ask about a styling issue, Claude automatically thinks in terms of your Vue.js + Tailwind stack. No re-explanation needed.

How It Works: Automatic Learning from Chat History

The magic behind Claude's Memory lies in its automatic learning mechanism. There's no complex setup — just enable it and let Claude do the work.

The Learning Process

  1. Extraction: Claude analyzes your chat history to identify important personal preferences and context
  2. Storage: This information is structured and stored securely in your Memory
  3. Application: When you start a new chat, Claude automatically references this Memory to provide better, more personalized responses

What Claude Learns

Claude's Memory captures information such as:

  • Your Role & Profession: Software engineer, product manager, content writer, etc.
  • Skill Level: Beginner, intermediate, expert (for various domains)
  • Technical Preferences: Programming languages, frameworks, tools you regularly use
  • Work Style: Preference for detailed explanations vs. concise bullets
  • Interests & Expertise: Your areas of focus and passion
  • Communication Style: Formal vs. casual, technical depth preference
  • Current Projects: Active projects and their requirements
  • Language & Localization: Your preferred language and timezone

Free for Everyone — A Major Shift

ℹ️
**Game-changing news:** Memory was previously exclusive to Claude Pro subscribers. In March 2026, Anthropic opened this feature to every single Claude user at no additional cost.

All of these users now have access to Memory:

  • Claude Free users
  • Claude Pro subscribers
  • Claude Team members
  • Claude Enterprise users

Cost: Nothing. It's completely free.

This democratization is significant. A feature that previously required a paid subscription — providing sophisticated personalization and context awareness — is now available to everyone. This levels the playing field and makes Claude a more capable assistant for users at all levels.

Enabling and Configuring Memory

Step-by-Step Setup

Enabling Memory on claude.ai is straightforward:

  1. Log in to claude.ai - Access your account
  2. Open Settings - Click the menu icon (☰) in the top left
  3. Navigate to Memory - Find the Memory section
  4. Enable Memory - Toggle Memory to "On"
  5. Wait for initialization - Claude will scan your chat history (usually takes a few minutes)

That's it! Claude immediately begins learning from your past conversations.

Settings Navigation:
Account Settings
├── Profile
├── Appearance
├── Privacy & Security
└── Memory ← Enable here
💡
**Tip:** On first enable, Claude processes your entire chat history. This might take a few minutes depending on your account age. Grab a coffee and let it work!

Fine-tuning Memory Settings

After enabling, you have control over specifics:

  • Review Memories: See exactly what Claude has learned about you
  • Enable/Disable Categories: Toggle automatic learning for specific areas
  • Set Learning Preferences: Control what types of information Claude captures
  • Privacy Levels: Adjust sensitivity for different information types

Privacy & Control — Your Data, Your Rules

Privacy concerns are legitimate. Anthropic takes this seriously, and Memory is designed with transparency as a core principle.

Viewing Your Memories

Complete transparency is built in. You can always see exactly what Claude remembers:

  1. Go to SettingsMemory
  2. Click View Memories
  3. See all learned information organized by category

From here, you can:

  • Edit specific memories to correct them
  • Delete memories you don't want remembered
  • Add custom memories manually

Memory Deletion Options

You have granular control:

  • Delete Individual Memories: Remove specific learned items
  • Delete by Category: Clear all memories in one area (e.g., all work-related memories)
  • Full Reset: Completely clear all Memory data and start fresh
Example Memory Organization:
Personal Profile
├── Role: "Senior product manager at tech startup"
├── Skills: "Product strategy, SQL, data analysis"
└── Preferences: "Prefers data-driven decisions"

Work Context
├── Current Projects: "Building analytics dashboard"
├── Tech Stack: "React, Node.js, PostgreSQL"
└── Team: "Cross-functional team of 5"

Communication Style
├── Format: "Prefers bullet points and examples"
├── Depth: "Intermediate technical level"
└── Language: "English, occasional Spanish"
⚠️
**Important:** Memory is stored securely on Anthropic's servers and is tied to your account only. Anthropic does not share Memory data with third parties. Review their Privacy Policy for full details.

Practical Tips for Maximum Value

Let's look at how to get the most from Claude's Memory feature.

Tip 1: Establish Your Professional Context

In your first conversation after enabling Memory, clearly establish who you are:

"I'm a full-stack developer with 8 years of experience. I primarily work with JavaScript/TypeScript, React, and Node.js. I'm comfortable with DevOps and cloud infrastructure (AWS). I'm learning Rust and interested in systems programming."

Claude will use this context to calibrate its responses — no more generic answers that don't match your actual capabilities.

Tip 2: Communicate Your Skill Level

Different domains might have different proficiency levels:

"I'm an advanced user with Figma and UI design, but completely new to no-code automation tools. Please explain basics carefully when discussing automation."

Claude learns and applies this nuance, adjusting explanations accordingly.

Tip 3: Describe Your Communication Preferences

Explicitly state how you like to receive information:

"I learn best with concrete examples. Please include code snippets for technical topics. I prefer conversational explanations over formal documentation style."

Your future Claude conversations will automatically adopt this style.

Tip 4: Document Active Projects

Good Memory-building prompt:
"I'm currently working on a personal project — a
budget tracking app. Stack: Next.js 14 frontend,
Python FastAPI backend, PostgreSQL database.
I'm solo on this project and expect to launch in Q2."

Future technical questions about your project will now have full context.

Tip 5: Periodic Memory Maintenance

💡
**Best practice:** Review your Memory monthly. As your skills grow or roles change, update your Memory to reflect your current state. Delete outdated project information. This keeps Claude's understanding fresh and accurate.

Claude Memory vs. Projects Memory vs. CLAUDE.md

Since Claude has multiple memory-related features, let's clarify how they work together.

Memory (This Feature)

  • Scope: Your entire account
  • Coverage: All conversations
  • Learning Type: Automatic from chat history
  • Purpose: Personal preferences and background
  • Persistence: Long-term (continuously updated)

Projects Memory

  • Scope: Within a specific Project
  • Coverage: Only that Project's conversations
  • Learning Type: Manual (you explicitly add information)
  • Purpose: Project-specific context and details
  • Persistence: While Project is active

CLAUDE.md

  • Scope: Within a specific Project
  • Coverage: Used as reference document
  • Learning Type: Manual (you write it)
  • Purpose: Explicit system instructions and rules
  • Persistence: As long as file exists

How They Work Together

The Three-Layer Memory Stack:

Layer 1: Memory (Global)
└─ "I'm a backend engineer who prefers Python"

Layer 2: Project Memory (Project-specific)
└─ "This project uses Django, PostgreSQL, Redis"

Layer 3: CLAUDE.md (Explicit rules)
└─ "For this project, always prioritize
    security. Use type hints everywhere."

Result: Highly personalized, context-aware assistance
ℹ️
**The Best Approach:** Use all three in harmony. **Memory** learns your global preferences. **Project Memory** adds project-specific context. **CLAUDE.md** provides explicit rules. Together, they create the most powerful Claude experience possible.

Practical Use Cases

Use Case 1: Career Learner

You're learning Python as a career change. Tell Claude:

"I'm transitioning from marketing to data science. I have basic Python knowledge but I'm new to pandas, numpy, and data visualization libraries."

Claude will provide learning-appropriate explanations, gradually increasing complexity as it detects your growing understanding.

Use Case 2: Technical Writer

You write documentation for developers. Tell Claude:

"I write API documentation for engineers. I need to balance technical accuracy with accessibility. I prefer examples in both JavaScript and Python."

Your future documentation-related conversations will automatically include both language examples and appropriate technical depth.

Use Case 3: Freelance Designer

You work with multiple clients on varied design projects. Tell Claude:

"I'm a freelance designer working with Figma, Adobe XD, and Sketch. My clients range from startups to enterprise. I need design system guidance, accessibility compliance help, and design critique from a user-centered perspective."

Claude will remember your varied context and can seamlessly switch perspectives based on project needs.

Use Case 4: Parent & Side-Hustle Developer

You work on personal projects alongside family time. Tell Claude:

"I'm a parent with limited evening/weekend coding time. I work on personal projects in JavaScript/Vue.js. I need efficient, time-conscious advice that respects my constraints."

Claude will prioritize efficiency and practicality in its responses.

Best Practices

  1. Be Specific: The more detail you provide about yourself, the better Claude learns
  2. Update Regularly: As your skills and roles evolve, update your Memory
  3. Review Quarterly: Check what Claude has learned and make corrections as needed
  4. Use Projects for Specifics: Don't overload Memory; use Project Memory for project-specific context
  5. Maintain Privacy: Regularly review what's stored and delete sensitive information you no longer need
  6. Combine Features: Use Memory + Projects Memory + CLAUDE.md together for maximum power

Getting Started Today

Ready to experience a more personalized Claude? Here's your action plan:

  1. Enable Memory in Settings (takes 2 minutes)
  2. Wait for initialization (takes a few minutes)
  3. In your next conversation, start by telling Claude about yourself
  4. Notice the difference in how Claude responds across future conversations
  5. Monthly review: Check your memories and refine

Wrapping up

Claude's Memory represents a fundamental shift in how AI assistants work. Rather than a blank slate every conversation, Claude becomes genuinely persistent, learning your preferences and context over time.

The fact that this feature is free for all users makes it transformative. Whether you're a student, professional, hobbyist, or free-tier user, you now have access to sophisticated personalization that was previously exclusive.

The future of AI assistance is personalized, context-aware, and respectful of user privacy. Claude's Memory feature is a significant step in that direction.

Start today. Enable Memory, tell Claude about yourself, and experience how much more helpful a remembered AI can be.

Share

Thank You for Reading

Claude Lab is ad-free, supported entirely by members like you. We publish practical guides daily with implementation code, benchmarks, and production-ready patterns. If you've found it useful, we'd love to have you on board.

  • Copy-paste ready implementation code
  • New advanced guides published daily
  • $5/mo or $10 for lifetime access
View Membership →

If you found this article helpful, a small tip ($1.50) would mean a lot to us. Your support helps keep this site ad-free and covers server and hosting costs.

Related Articles

Claude.ai2026-07-12
Using Claude's Reflect Monthly Review to Tune a Solo Dev Rhythm
How to read Claude's new Reflect monthly review as a planning tool for the month ahead rather than a report card, grounded in the day-to-day rhythm of solo development.
Claude.ai2026-07-09
Claude Card Declined: A Complete Troubleshooting Guide for Pro, Max, and API Users
When Claude tells you 'Your card was declined,' the cause is rarely obvious from the error text alone. This guide separates the three layers where a decline actually happens — your issuing bank, Stripe's fraud engine, and Anthropic's account state — and walks you through fixes plus fallback payment methods that almost always get the charge through.
Claude.ai2026-07-06
When Instructions Fade in Long Conversations: Measuring Adherence and Pulling It Back
System-prompt instructions quietly lose their grip as a conversation grows. This piece turns that drift into a number you can track, then fixes it with prompt structure and mid-conversation re-anchoring, backed by runnable code and measured results.
📚RECOMMENDED BOOKS
Build a Large Language Model (From Scratch)
Sebastian Raschka
LLM Dev
Prompt Engineering for LLMs
Berryman & Ziegler
Prompting
AI Engineering
Chip Huyen
AI Eng
* Contains affiliate links
See all →