squish
Verified Safeby michielhdoteth
Overview
Local-first persistent memory for Claude Code, enabling auto-capture, folder context, privacy filtering, and optional team collaboration.
Installation
node dist/index.jsEnvironment Variables
- DATABASE_URL
- REDIS_URL
- SQUISH_EMBEDDINGS_PROVIDER
- SQUISH_OPENAI_API_KEY
- SQUISH_OLLAMA_URL
- SQUISH_WEB_PORT
Security Notes
The server's web UI runs on localhost, limiting direct network exposure. Team mode utilizes PostgreSQL and Redis, which would require careful network configuration if exposed externally, though the provided Docker Compose binds them locally. The `secret-detector.ts` effectively identifies sensitive patterns, and the privacy features allow filtering and redaction. No 'eval' or obvious obfuscation detected. Installer scripts download from GitHub releases, relying on GitHub's integrity. Default PostgreSQL credentials in docker-compose.yml (`squish:squish_dev`) are only for development and should be changed for production deployments.
Similar Servers
context-sync
Context Sync provides AI systems with persistent, queryable memory across all development tools, sessions, and projects, allowing AI to remember codebase details, architectural decisions, and conversation history.
mcp-memory-keeper
Provides persistent context management for Claude AI coding assistants, ensuring work history, decisions, and progress are preserved across sessions and context limits.
claude-memory-mcp
Provides local, persistent, searchable memory for Claude Desktop and other MCP-compatible AI assistants.
memcord
Memcord is a privacy-first, self-hosted MCP server designed to organize chat history, summarize messages, and enable AI-powered search across past conversations, keeping all data secure and under user control.