memory-journal-mcp
by neverinfamous·★ 16·Score 42
An MCP server providing AI memory with project detection, GitHub integration, and knowledge graphs for context-aware development.
Overview
Memory Journal MCP Server is a comprehensive tool designed to solve AI session amnesia by providing persistent memory and context awareness across development workflows. It offers 70 MCP tools for managing project history, bridging disconnected AI sessions, and integrating with GitHub workflows. The server implements session intelligence, dynamic project routing, and hybrid search capabilities combining FTS5 keywords with semantic vector similarity. Its standout features include the Hush Protocol for structured team communication, knowledge graphs with Mermaid visualization, and Code Mode for efficient multi-step operations with significant token savings.
Try asking AI
After installing, here are 7 things you can ask your AI assistant:
When to choose this
Choose this MCP server if you're working on GitHub projects that need persistent memory between AI sessions and require deep GitHub integration for workflow automation and team collaboration.
When NOT to choose this
Avoid if you need a lightweight solution or are not working with GitHub, as this server is feature-rich but may be overkill for simpler use cases or projects not using GitHub.
Tools this server exposes
12 tools extracted from the READMEcreate_entryCreate a new journal entry with content, tags, and optional metadata
search_entriesSearch through journal entries using keywords, tags, and filters
semantic_searchPerform semantic search using vector similarity to find related entries
list_github_issuesList GitHub issues with filtering and pagination options
create_github_issueCreate a new GitHub issue with title, body, and labels
create_github_pull_requestCreate a new GitHub pull request with changes from a branch
get_github_repository_infoRetrieve comprehensive information about a GitHub repository
link_entriesCreate relationships between different journal entries
generate_knowledge_graphCreate a visual graph showing relationships between entries and entities
export_to_markdownExport journal entries or search results to Markdown format
execute_codeExecute code in a secure sandbox environment for complex operations
create_session_summaryGenerate a summary of work done in the current AI session
Comparable tools
Installation
Installation
npm Package
npm install -g memory-journal-mcpDocker
docker pull writenotenow/memory-journal-mcpClaude Desktop Configuration
Add to claude_desktop_config.json:
{
"mcpServers": {
"memory-journal": {
"command": "npx",
"args": ["-y", "memory-journal-mcp"]
}
}
}Custom Tool Filtering
Set environment variable: MEMORY_JOURNAL_MCP_TOOL_FILTER=starter or use CLI --tool-filter essential
FAQ
- How does the session briefing work?
- At the start of each new AI session, agents automatically read the 'memory://briefing' resource which contains project history, active flags, GitHub insights, and latest session summaries to provide context continuity.
- Can I filter tools to reduce server overhead?
- Yes, use tool filters like 'starter' (~11 tools), 'essential' (~7 tools), or specify custom selections with --tool-filter flag to only expose necessary tools.
Compare memory-journal-mcp with
Last updated · Auto-generated from public README + GitHub signals.