$ loading_
Give Claude Code persistent cross-session memory for project context and preferences.
This MCP tool appears to persist cross-session memory locally in a Markdown file inside the repository, with no declared secrets or remote endpoints. Based on the available materials, the overall risk is relatively low, but it should still be used in a constrained environment because it runs as a local MCP service, writes repository files, and has weak community adoption and unclear maintenance.
The materials explicitly state that no keys or environment variables are required. No API tokens, account credentials, or other sensitive authentication requirements are described, so credential exposure and abuse risk appears low.
The materials declare no remote endpoints, and the description only mentions persisting memory to a local .claude-memory.md file in the repository. There is no evidence of user data being transmitted to external services.
The system checks indicate that it executes code. As an MCP server, it would typically run a local process to provide its functionality. This is a normal capability for this class of tool; the current materials do not show dangerous privileges beyond its stated purpose, but it should still be treated cautiously as a local executable component.
The description states that it reads and writes a .claude-memory.md file in the repository to persist cross-session memory, so it at least has access to and can modify data in the repo working directory. The available materials do not indicate broader data access, but its accessible directory scope should still be restricted.
The project is open source and therefore auditable in principle, which is a positive factor that lowers risk. However, it comes from a third-party registry, lacks a README, has no declared license, shows 0 stars, and has unknown maintenance status, indicating weak supply-chain maturity and limited trust signals; code and dependencies should be reviewed before use.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "claude-memory-mcp" yet — see the docs or source repo.
Please remember and store these repo conventions: use TypeScript, run tests before commits, return API fields in camelCase, and place all new components under src/components.
The tool persists the conventions to .claude-memory.md so future sessions can reuse them.
Read the current repo memory and tell me where we left off, what the outstanding tasks are, and what to do next.
Returns a progress summary, open tasks list, and recommended next steps based on the memory file.
Save this debugging result to memory: login failures were caused by a missing JWT_SECRET environment variable; the temporary fix is adding it to .env.local, and a required check should be added to CI later.
The tool stores the root cause, temporary fix, and follow-up action for future troubleshooting and collaboration.
Give Claude Code persistent memory, checkpoints, and project sync across sessions.
Persist Claude Code conversations and retrieve relevant context across sessions.
Give Claude Code persistent memory and semantic context retrieval across sessions.
Query Claude Code project memory, history, and preferences across projects.
Share project memory files over MCP across Claude Code and Claude.ai.
Provide shared memory for Claude across apps with sessions, handoffs, and artifacts.