Can I manually manage what Claude remembers?

Yes, you can use slash commands like /remember to add information, /forget to remove it, and /memories to list everything Claude currently knows about the project.

How does Claude store project memories?

Memories are stored locally in a JSON file at {baseDir}/.memories/project_memory.json, ensuring your context remains private and persistent within your project directory.

Does stored memory override my current instructions?

No, the skill is designed to prioritize current explicit user requests over stored memories, allowing you to override past decisions at any time.

What happens if the memory file is missing?

If the memory file is not found, the skill will initialize a new memory file in the standard location and prompt you to set up memory persistence.

Persistent Project Memory

Name: Persistent Project Memory
Author: BbgnsurfTech

byBbgnsurfTech

•

Productivity & Workflow

Maintains long-term context across development sessions by storing and retrieving architectural decisions, coding conventions, and user preferences.

About

The Persistent Project Memory skill empowers Claude to maintain continuity across multiple development sessions by leveraging a local storage system for project-specific context. By automatically tracking architectural decisions, coding conventions, and tool selections, this skill ensures that Claude provides consistent assistance that aligns with historical choices. It eliminates the need for repetitive context-setting by allowing the AI to recall past discussions and established project patterns through simple natural language triggers or dedicated slash commands, making Claude a more effective long-term collaborator on complex codebases.

Key Features

3 GitHub stars
Manual memory management via /remember and /forget slash commands
Intelligent conflict resolution prioritizing current user instructions
Persistent tracking of architectural patterns and library choices
Silent integration that informs responses without intrusive announcements
Automated context retrieval from local JSON storage

Use Cases

Maintaining consistent architectural patterns and library choices across a multi-week project
Recalling rationale for past technical decisions to ensure future changes remain compatible
Enforcing project-specific coding styles and naming conventions without repeated prompting

About

Key Features

3 GitHub stars
Manual memory management via /remember and /forget slash commands
Intelligent conflict resolution prioritizing current user instructions
Persistent tracking of architectural patterns and library choices
Silent integration that informs responses without intrusive announcements
Automated context retrieval from local JSON storage

Use Cases

Maintaining consistent architectural patterns and library choices across a multi-week project
Recalling rationale for past technical decisions to ensure future changes remain compatible
Enforcing project-specific coding styles and naming conventions without repeated prompting