RAWThink runs on your own machine — the client starts it, so there is no endpoint to ping. 100 installs a week from pypi. Last commit 31 Jul 2026.
Persistent memory for AI thinking partnerships: hybrid search and a temporal knowledge graph.
Nothing serious here today
Today is the operative word: we check RAWThink every 15 minutes and re-read its code on every release. Watch it and you find out the day that stops being true.
Connect this server
This server runs on your own machine — install it with the package manager and the client starts it for you. Package name taken from the official registry entry.
claude mcp add rawthink-mcp -- uvx rawthink-mcp{
"mcpServers": {
"rawthink-mcp": {
"command": "uvx",
"args": [
"rawthink-mcp"
]
}
}
}[mcp_servers.rawthink-mcp]
command = "uvx"
args = ["rawthink-mcp"]{
"mcpServers": {
"rawthink-mcp": {
"command": "uvx",
"args": [
"rawthink-mcp"
]
}
}
}{
"servers": {
"rawthink-mcp": {
"type": "stdio",
"command": "uvx",
"args": [
"rawthink-mcp"
]
}
}
}Optional settings from the registry entry, left out of the block above:
RAWTHINK_VAULT (Path to the vault directory holding sessions, qnotes and the knowledge graph), MEMORY_FILE_PATH (Path to memory.jsonl. Defaults to <vault>/memory.jsonl), QDRANT_URL (Qdrant server URL), QDRANT_PATH (Set to a directory to run Qdrant embedded instead of via Docker), OLLAMA_URL (Ollama server URL. Without it, retrieval degrades to sparse-only), OLLAMA_MODEL (Embedding model. Changing it requires a full reindex), RAWTHINK_TOOL_PROFILE (Which tools to expose: recall (4, read-only), record (5, the write path), or full (17)), RAWTHINK_BM25_STATE (Path to the BM25 state file. Defaults to <vault>/.bm25_state.json), RAWTHINK_LOG_LEVEL (Log level for diagnostics on stderr).
Alternatives to RAWThink
same job, measured the same waySelf-hosted persistent memory for AI agents: knowledge graph and hybrid search.
Persistent memory for AI tools. Local-first knowledge graph with hybrid search.
Persistent knowledge memory for AI agents. Hybrid search, code graph, pgvector.
Persistent memory and knowledge graphs for AI agents. Hybrid search, context checkpoints, and more.
Persistent long-term memory for AI agents: semantic search, knowledge graph, and task canvas.
Persistent memory for AI assistants with semantic search and knowledge graph relationships.
Persistent memory infrastructure for AI agents. Identity, knowledge graph, and search.
Persistent memory, hybrid search and a goal graph for AI agents, over stdio or remote HTTP.
RAWThink — questions
Answers built from our own checks of this server.