Data Cleaner Agent runs on your own machine — the client starts it, so there is no endpoint to ping. 162 installs a week from pypi. Last commit 10 Aug 2026.
Clean messy CSVs: an LLM picks the steps, tested code does the work. Data untouched by the model.
Today is the operative word: we check Data Cleaner Agent every 15 minutes and re-read its code on every release. Watch it and you find out the day that stops being true.
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 data-cleaner-agent -- uvx agentic-csv-cleaner
{
"mcpServers": {
"data-cleaner-agent": {
"args": [
"agentic-csv-cleaner"
],
"command": "uvx"
}
}
}
[mcp_servers.data-cleaner-agent]
command = "uvx"
args = ["agentic-csv-cleaner"]
{
"mcpServers": {
"data-cleaner-agent": {
"args": [
"agentic-csv-cleaner"
],
"command": "uvx"
}
}
}
{
"mcpServers": {
"data-cleaner-agent": {
"args": [
"agentic-csv-cleaner"
],
"command": "uvx"
}
}
}
Local LLM does Claude's bulk work: reads logs and files, researches the web, writes test-gated code.
Finds the LLM model IDs pinned in your code that vendors have already switched off, or have dated to
Find the LLM/AI calls you didn't need — where plain code or a regex does the job. No model.
Detect prompt injection, jailbreaks, and code injection in untrusted text before it reaches an LLM.
Check whether a Claude Code skill was tested and whether it works before installing it.
Turn long URLs into shre.me/code, and read the clicks: day, country, device, referrer.
Delegate bulk code edits to a cheap worker model. Gated, atomic. PHP and Python.
Code changes and unit tests by local MLX models, behind a gate a machine can run.
Answers built from our own checks of this server.