convert_tempLABSCO SUMMARY
The five-second jobs stop breaking your concentration.
Hashing a file, decoding a base64 blob, generating a password, checking a word count — each of these normally means a terminal or a website, and here they are answered inside whatever you were already doing.
What it is
A set of everyday utilities built on FastMCP: file operations, hashing, encoding, text statistics, date formatting, temperature conversion and password generation. Everything runs locally over stdio.
What you get
- Text files read in full, written with the directory created if it does not exist, and directories listed with file sizes
- MD5, SHA1 and SHA256 hashes of either a string or a whole file's contents
- Base64 encoding and decoding
- Word, character and line counts, with averages per line and per word
- The current date and time, broken down into parts or formatted with a strftime pattern
- Passwords generated to a length you set, with uppercase, lowercase, digits and symbols each switchable
- Celsius and Fahrenheit converted to two decimal places
Requirements
Python 3.8 or newer with the dependencies installed. It reads and writes files on the machine it runs on, using that user's permissions, so point it at directories you are content for it to work in.
Setup effort
Build from source — clone the repository, then follow the steps on this page and add the configuration for your client
USE CASES
01
Verifying a downloaded file
Hash the file with SHA256 and compare it against the published checksum, without opening a shell to do it.
02
Reading what a base64 blob actually contains
Decode it in place — the fastest way to find out whether a token, a config value or a payload is what it claims to be.
03
Producing a password to a policy
Set the length and switch the character classes on or off to match whatever the system in front of you demands, rather than fighting a generator's defaults.
04
Checking whether a draft fits a limit
Count words, characters and lines, with the averages alongside, before you find out from a form that it is too long.
05
Writing a file into a directory that does not exist yet
The write creates the directory on the way, which turns a two-step task into one and avoids a failure you would otherwise have to read.
06
Formatting a timestamp the way something else expects it
Get the current date and time in a strftime pattern of your choosing, or as separate parts when you need to assemble it yourself.
9 tools
Listed in project docs. Your client may show a different list.
read_filewrite_filelist_directorycalculate_hashbase64_encode_decodeget_datetime_infotext_statsgenerate_passwordBEFORE YOU INSTALL
What to know first
Tools you get
9 tools named in the project's documentation. read_file, list_directory and write_file act with the Python process's rights; write_file writes anywhere and creates directories. No allow-list or sandbox setting limits them.
Requires
Python 3.8 or higher
Where it runs
Both documented clients run the server on your machine.
INSTALLATION
Connect your client
Clone the repository, then follow the steps on this page and add the configuration for your client. This one is built from source rather than installed with a single command.
Server setup
TERMINAL
git clone https://github.com/aiforhumans/local-utils-mcp.git
cd local-utils-mcp
python -m venv .venv
source .venv/bin/activateOn Windows use .venv\Scripts\activate instead.
TERMINAL
pip install -r requirements.txt
python server.pyMCP Clients
LM Studio
LM STUDIO · JSON
{
"mcpServers": {
"local-utils": {
"command": "<YOUR_PATH_TO_LOCAL_UTILS_MCP>/.venv/bin/python",
"args": [
"<YOUR_PATH_TO_LOCAL_UTILS_MCP>/server.py"
],
"env": {}
}
}
}Claude Desktop
CLAUDE DESKTOP · JSON
{
"mcpServers": {
"local-utils": {
"command": "<YOUR_PATH_TO_LOCAL_UTILS_MCP>/.venv/bin/python",
"args": [
"<YOUR_PATH_TO_LOCAL_UTILS_MCP>/server.py"
]
}
}
}
