linux

HOW TO : Run Anthropic Computer Use Tool on a Windows Machine

Anthropic released their new Claude Sonnet 3.5 model yesterday that has a new capability to control computers. Computer Use capability allows Claude to directly interact with computer interfaces, enabling tasks like web browsing, data analysis, and file manipulation – all through natural language instructions. Similar to tools, but now you don’t have to define specific tools. I think this opens up a whole new window of opportunities to leverage LLMs for.

Anthropic shared a quick start guide to run the model in a container, but the instructions are for Mac/Linux based workstations. I had to make some tweaks to run them on a windows workstation.

Documenting them for anyone that might be trying to do the same

  • Install Docker Desktop
  • Open a command prompt
  • Run the following command to set your anthropic api key system variable
    • set ANTHROPIC_API_KEY=YOUR-ANTHROPIC-KEY
  • Run the following command to start the docker container
    • docker run -e ANTHROPIC_API_KEY=%ANTHROPIC_API_KEY% -v $HOME/.anthropic:/home/computeruse/.anthropic -p 5900:5900 -p 8501:8501 -p 6080:6080 -p 8080:8080 -it ghcr.io/anthropics/anthropic-quickstarts:computer-use-demo-latest
  • Launch the streamlit app by opening this URL in your browser http://localhost:8080/

HOWTO : Bulk deletes in vi

Use “dG” command, if you want to delete all lines in a file starting from the line under the cursor in vi.

Additional commands to delete lines

  1. dd deletes the whole line under the cursor.
  2. xdd deletes multiple (x) lines, starting at the cursor. For example 10dd deletes 10 lines
  3. d$ deletes to the end of the line, starting at the cursor.

HOW TO : Search which package contains a filename

If you are using a Linux system that uses yum for package management (like Fedora, Centos, RHEL), you can use the following command to find out which package contains a file. This is useful when you want to figure out which package to install. For example, dig (DNS utility) doesn’t come pre-installed on the system. And running “sudo yum install dig” doesn’t do anything.

sudo yum whatprovides '*/dig'

This returns

Loaded plugins: fastestmirror
 Loading mirror speeds from cached hostfile
 32:bind-utils-9.8.2-0.47.rc1.el6.x86_64 : Utilities for querying DNS name servers
 Repo : base
 Matched from:
 Filename : /usr/bin/dig

breaking down the command options

whatprovides : Is used to find out which package provides some feature or file. Just use a specific name or a file-glob-syntax wildcards to list the packages available or installed that provide that feature or file.

HOW TO : Enable wildcard domains in Squid

We were trying to modify some ACL (access control lists) in squid to allow traffic to certain websites. Instead of adding each individual hostnames in a domain, we wanted to add all traffic to a certain domain.

Document on the interwebs is old or not clear on how to achieve this.

After some trial and error, here is what works

say you want to allow all traffic to the google.com domain, you create a access list using dstdomain like below

acl name_of_acl dstdomain .google.com

The “.” before the domain name acts as a wildcard

Then you use the acl to allow http access to it like below

http_access allow name_of_acl

HOW TO : Use awk to print values larger than certain number

Quick how to on using awk to filter results if a certain value (column) is larger than a set value.

For example, if you have a file (servers.txt) with lines in this format

a_datacenter, servers 20
 error, servers xyz
 b_datacenter, servers 21
 c_datacenter, servers 50

and you want to show only the lines that have server value larger than 20, you can do this in awk by running

grep datacenter servers.txt | awk '$3 > 20  {print ;}' | more

breaking down the commands

grep – parsing down the output to just show the lines containing datacenter

awk – $3 > 20 : Get the third variable (awk seperates text using spaces by default) and check if it is greater than 20

print – print the entire line

HOW TO : Find files, search for content in them, replace the content

The title pretty much says it all :). Here is a quick  one liner, using multiple tools, to look for files in a directory, search for certain content in them and replace them with other content

[code]find -type f | xargs grep -l ORIGINAL_CONTENT | xargs perl -p -i -e ‘s/ORIGINAL_CONTENT/NEW_CONTENT/g’ [/code]

You can theoretically take out the grep (second command) and directly pipe the find output to perl and get the same outcome.

Going over list of the options used

find

  • “-type f” lists all objects of type file in the directory (and sub directories)

grep

  • “-l” lists the names of the files (with relative path) which have the text ORIGINAL_CONTENT in them

perl

  • “-p” forces perl to loop through requests. In this case files
  • “-e” tells perl that the next argument is a perl statement
  • “-i” tells perls to edit the file in place (i.e. no need for an output file)