Production-ready MCP memory server with semantic search, project conventions learning, and comprehensive memory management for OpenCode
Wellknown found it in public sources; nobody has proven control of it yet. Claiming takes one click if the repository is under your GitHub account, or a small file on your domain otherwise. Verified owners get the badge, 15-minute checks, status alerts, edits that outrank crawled data, and a ranking boost.
Agents can do it too: POST https://wellknown.network/api/v1/claims with {"agent":"opencode-mcp-memory","method":"well_known_file"} — machine-readable steps at claim.json, guide at /docs/claim.
Everything here was measured by our prober or read from a registry. Nothing is self-reported.
Attributed to the source that supplied each field. Treated as claims, not facts.
# OpenCode MCP Memory A production-ready MCP (Model Context Protocol) memory server for OpenCode with semantic search, project conventions learning, and comprehensive memory management. ## Features - **Semantic memory search** via sentence-transformers with automatic text search fallback - **Multiple memory types**: Working, semantic, episodic, procedural, plus domain-specific types - **Project convention learning**: Auto-detect project type, tools, commands, dependencies - **Knowledge graph**: Automatic relationship detection between related memories - **Duplicate detection**: MD5-based content hashing to prevent redundant storage - **Configurable embeddings**: Support for different sentence-transformers models - **Full-text search**: SQLite FTS5 with fallback LIKE queries - **Comprehensive metrics**: Track operation performance and success rates - **Input validation & security**: Prevent path traversal and malicious inputs ## Quick Start After installation, run the server directly: ```bash mcp-memory ``` Running from a source checkout without installing uses the launcher, which creates a `.venv` and installs dependencies on first run: ```bash python mcp_memory/launcher.py ``` ## Installation ### From PyPI (recommended) ```bash pip install opencode-mcp-memory ``` Then run directly: ```bash mcp-memory ``` ### From Source (development) ```bash # Clone and install in editable mode git clone https://github.com/opencode/mcp-memory.git cd mcp-memory pip install -e . # Run the server mcp-memory ``` ### For OpenCode Integration Add the server to your opencode config (`opencode.json` or `opencode.jsonc`): ```jsonc { "mcp": { "opencode-memory": { "type": "local", "command": ["mcp-memory"] } } } ``` ## Architecture - `server.py` - FastMCP server with 18 tools - `launcher.py` - Venv-aware launcher for running from source without installing - `memory.py` - Memo…
Mapped onto the structured taxonomy from declared text and observed tool names. Confidence shown for derived entries.
Every source is kept verbatim. Field changes are logged as events.