$ loading_
Share, search, and reuse local memory across multiple AI coding agents.
This tool is described as a local stdio MCP server with no API keys and no external services, so overall risk appears relatively low. However, it inherently executes locally and can read/write under ~/.claude/memory, so it should be used with awareness of its local data impact.
The materials explicitly state that no keys or environment variables are required. No API tokens, account passwords, or other sensitive credentials are requested, so credential exposure appears minimal.
It is described as a local stdio MCP server with no remote endpoints and no external services. Based on the provided materials, there is no indication that user data is transmitted over the network or to third-party hosts.
The system flags it as executes-code, meaning it runs a local service process as an MCP tool. This is a normal capability for such tools, and the materials do not show red flags such as privilege escalation or unrelated high-privilege system actions.
The description explicitly states read/search/write access over ~/.claude/memory, meaning it can read, search, and modify shared memory data in that local directory. This matches its stated purpose, and there is no evidence of broader filesystem access requests.
Positive factors include being open source under the MIT license, which makes source review possible. However, it comes from a third-party registry, has 0 GitHub stars, unknown maintenance status, and no README, so supply-chain trust is limited and source/dependency review is advisable before installation.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "memory-mcp" yet — see the docs or source repo.
Write the current project's tech stack, folder conventions, and key development rules into memory-mcp so I can reuse them later in Claude Code, Cursor, and Codex.
The tool stores project context in local shared memory for later reading and search by different agents.
Search memory-mcp for past notes about "authentication approach" or "database selection" and summarize the technical decisions made earlier.
It returns relevant memory entries and summarizes reusable past decisions.
Save the cause, fix steps, and cautions from this build failure investigation into memory-mcp so similar issues can be found quickly later.
The debugging learnings are saved as searchable memory for future troubleshooting.
Provide shared memory for Claude across apps with sessions, handoffs, and artifacts.
Persist Claude Code conversations and retrieve relevant context across sessions.
Share project memory files over MCP across Claude Code and Claude.ai.
Provide persistent local memory and semantic search for MCP AI clients.
Provide persistent local memory for development context, decisions, and past discussions.
Give Claude Code persistent memory and semantic context retrieval across sessions.