Tokenmizer runs on your own machine — the client starts it, so there is no endpoint to ping. 150 installs a week from pypi. Last commit 23 Oct 2026.
Graph-backed session memory for LLMs: checkpoint, resume, file analysis, token-savings stats.
We read the source, 17 h ago · rules 995398eba69f
What this server is able to do. For an MCP server this is often the job itself — a terminal server runs commands because that is what it is for. Listed so you know what you are plugging in, not as an accusation.
r'\.npmrc|\.nvmrc|\.editorconfig|\.bashrc|\.zshrc|\.babelrc|'
# Pre-download tiktoken's BPE vocabulary at BUILD time. # # tiktoken ships no vocabulary — it fetches one from # openaipublic.blob.core.windows.net the first time an encoding is # used. Token counting is on the hot path of every proxied request, so # leaving that as a runtime …
Is this your server and something here is wrong? Tell us — corrections are free and do not require a plan.
We found places where it runs commands, builds paths or queries from values it is given. None of that is a flaw by itself — it becomes one when the code changes, and code changes quietly between releases. We re-read it on every one.
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 tokenmizer -- uvx tokenmizer
{
"mcpServers": {
"tokenmizer": {
"command": "uvx",
"args": [
"tokenmizer"
]
}
}
}
[mcp_servers.tokenmizer]
command = "uvx"
args = ["tokenmizer"]
{
"mcpServers": {
"tokenmizer": {
"command": "uvx",
"args": [
"tokenmizer"
]
}
}
}
{
"servers": {
"tokenmizer": {
"type": "stdio",
"command": "uvx",
"args": [
"tokenmizer"
]
}
}
}
Optional settings from the registry entry, left out of the block above:
TOKENMIZER_URL (Base URL of the running TokenMizer proxy (default http://localhost:8000)), TOKENMIZER_API_KEY (API key if the proxy has auth enabled).
MCP server that checkpoints long agent tasks: roll back to any step, resume in a new session.
Session checkpoint & recovery for Claude Code: resume, search, tag, and export recorded sessions.
Persistent graph-backed conversational memory for AI agents.
Checkpoint and resume for AI agent runs, so crashed runs pick up where they left off.
Evidence-backed project memory and checkpoints for AI coding agents.
Stock research memory for Claude: theses, sourced claims and SEC filing alerts in a local graph
SQLite-backed MCP server for persistent memory, full-text retrieval, and graph traversal.
Recall, check, search, timeline, and graph tools for fiction writers via Seizn Memory v3.
Answers built from our own checks of this server.