Lesson 12: Advanced Topics and 20 Essential FAQs

⏱ Est. reading time: 3 min Updated on 5/7/2026

12.1 Advanced: Auto Dream (Dark Launch Feature)

Auto Dream acts as the "overnight organizer" for Claude Code, analogous to human sleep. It periodically reviews and compresses auto memory:

  • Trigger: Occurs after 24 hours and more than 5 sessions.
  • Function: Merges duplicates, converts relative dates to absolute, and prunes stale rules.
  • Status: Currently in Dark Launch mode, controlled by server-side flags. If your /memory UI doesn't show it, you can achieve similar results by telling the LLM: "Help me organize my auto memory."

12.2 Troubleshooting Checklist

Issue Verification Steps
auto memory not recording This is normal. It's not a recorder; it requires LLM judgment. Try manually saying "Note: xxx" to trigger it.
claude-mem errors Check ~/.claude-mem/logs/. Ensure the Bun Runtime is installed.
Session start is slow Check if MEMORY.md is too large (over 200 lines). Check for Hook timeouts.
Can't find info across projects Ensure you are using the mem-search skill specifically, not just a standard search.

12.3 20 Core Q&As (Selected)

Q1: Are auto memory and CLAUDE.md the same thing? No. CLAUDE.md consists of human-written hard rules; auto memory consists of LLM-written autonomous notes.

Q2: Where does claude-mem send my code? It stores data locally (SQLite/Chroma), but the "distillation" process sends snippets to the Anthropic API for summarization.

Q3: How do I completely uninstall claude-mem? Run /plugin uninstall and delete the ~/.claude-mem folder.

Q4: Can auto memory be shared across projects? No, it is limited to a single Working Tree. For cross-project sharing, use claude-mem.

Q5: Is more memory always better? No. Excessive entries dilute signals and increase token costs. Maintaining a "high signal-to-noise ratio" in your memory is best practice.

Q6: Can I manually edit memory files? Yes, they are plain Markdown files or standard SQLite databases.

Q7: How do memory systems handle sensitive info? Currently, systems do not automatically filter secrets. It is recommended to perform regular audits and use --bare mode in sensitive projects.

(Note: Please refer to the original full document for more Q&As.)

Related Tools & Resources

Skill Marketplaces

M

Matt Pocock's AI Skills

A collection of small, adaptable, and composable AI agent skills designed to resolve common development issues, enhance agent-developer collaboration, and improve code quality.
A

Agent Skills

Production-grade engineering skills for AI coding agents, encoding workflows, quality gates, and best practices for software development.

Recommended Plugins

🏗️

Feature Dev

Comprehensive feature development workflow with a structured 7-phase approach. Agents: code-explorer, code-architect, code-reviewer.

Related Products

📦

mempalace

MemPalace is a local-first AI memory tool designed to store conversation history verbatim and retrieve it through semantic search, without summarizing or paraphrasing. It utilizes a structured index, categorizing information into "wings," "rooms," and "drawers" to enable scoped searches. The retrieval layer is pluggable (ChromaDB by default), ensuring all data remains on the user's machine unless opted out. It achieves an impressive 96.6% R@5 raw recall on the LongMemEval benchmark, notably without requiring LLMs or API calls, showcasing high performance in a local environment.
📦

LangGraph

LangGraph, from LangChain, Inc., is a powerful low-level orchestration framework engineered for building, managing, and deploying sophisticated, long-running, and stateful multi-agent applications. It represents application logic as cyclic Directed Acyclic Graphs (DAGs), giving developers precise control over LLM execution, including when to pause, loop, or transfer control. LangGraph ensures durable execution, seamless human-in-the-loop capabilities, and comprehensive short-term and long-term memory management. Deep integration with LangSmith provides robust debugging and production-ready deployment, making it the ideal choice for developing complex and reliable AI agents.