The Bash Instruct Family
Collection
The bash instruct series of datasets • 3 items • Updated
Datasets:
messages listlengths 3 3 | category stringclasses 3
values | utility stringclasses 89
values |
|---|---|---|
[
{
"role": "system",
"content": "You are a Bash expert."
},
{
"role": "user",
"content": "Give 'dd if=/dev/zero of=test.img bs=1M count=1024' a boosted CPU priority of -5."
},
{
"role": "assistant",
"content": "sudo nice -n -5 dd if=/dev/zero of=test.img bs=1M count=1024"
}
] | single | nice |
[
{
"role": "system",
"content": "You are a helpful assistant that writes correct, idiomatic Bash commands."
},
{
"role": "user",
"content": "Show the kernel routing table numerically."
},
{
"role": "assistant",
"content": "netstat -rn"
}
] | single | netstat |
[
{
"role": "system",
"content": "You are a helpful assistant that writes correct, idiomatic Bash commands."
},
{
"role": "user",
"content": "Please show its size, owner and permissions in notes.py."
},
{
"role": "assistant",
"content": "stat notes.py"
}
] | single | stat |
[
{
"role": "system",
"content": "You translate natural-language requests into correct Bash commands."
},
{
"role": "user",
"content": "Check if invoice.py exists and cat it, otherwise say it's missing."
},
{
"role": "assistant",
"content": "[ -f invoice.py ] && cat invoice.py || echo ... | pipeline | chain |
[
{
"role": "system",
"content": "You are a Linux shell expert. Respond with Bash only."
},
{
"role": "user",
"content": "Recursively make everything in data readable and writable by the owner only."
},
{
"role": "assistant",
"content": "chmod -R 600 data"
}
] | single | chmod |
[
{
"role": "system",
"content": "You are a helpful assistant that writes correct, idiomatic Bash commands."
},
{
"role": "user",
"content": "List logged-in users without the header line."
},
{
"role": "assistant",
"content": "w -h"
}
] | single | w |
[
{
"role": "system",
"content": "You are a command-line assistant specializing in Bash and GNU coreutils."
},
{
"role": "user",
"content": "Show kernel notice-level messages."
},
{
"role": "assistant",
"content": "dmesg -l notice"
}
] | single | dmesg |
[
{
"role": "system",
"content": "You are a command-line assistant specializing in Bash and GNU coreutils."
},
{
"role": "user",
"content": "Read the hostname straight from its config file."
},
{
"role": "assistant",
"content": "cat /etc/hostname"
}
] | single | hostname |
[
{
"role": "system",
"content": "You are a Bash expert."
},
{
"role": "user",
"content": "Read manifest.ini, then count the lines — as a pipeline."
},
{
"role": "assistant",
"content": "cat manifest.ini | wc -l"
}
] | pipeline | wc |
[
{
"role": "system",
"content": "You are a helpful assistant that writes correct, idiomatic Bash commands."
},
{
"role": "user",
"content": "Make a timestamped log directory under logs/."
},
{
"role": "assistant",
"content": "mkdir -p logs/$(date +%Y-%m-%d)"
}
] | pipeline | cmdsub |
[
{
"role": "system",
"content": "You are a helpful assistant that writes correct, idiomatic Bash commands."
},
{
"role": "user",
"content": "Start 'dd if=/dev/zero of=test.img bs=1M count=1024' with nice value 10 so it doesn't hog the CPU."
},
{
"role": "assistant",
"content": "nice -... | single | nice |
[
{
"role": "system",
"content": "You are a command-line assistant specializing in Bash and GNU coreutils."
},
{
"role": "user",
"content": "Display w with numeric IPs."
},
{
"role": "assistant",
"content": "w -i"
}
] | single | w |
[
{
"role": "system",
"content": "You are a helpful assistant that writes correct, idiomatic Bash commands."
},
{
"role": "user",
"content": "Given index.log, count each unique line, most common first."
},
{
"role": "assistant",
"content": "cat index.log | sort | uniq -c | sort -nr"
... | pipeline | uniq |
[
{
"role": "system",
"content": "You are a Bash expert."
},
{
"role": "user",
"content": "List all open TCP connections."
},
{
"role": "assistant",
"content": "lsof -i tcp"
}
] | single | lsof |
[
{
"role": "system",
"content": "You are a helpful assistant that writes correct, idiomatic Bash commands."
},
{
"role": "user",
"content": "Give './import.sh' a boosted CPU priority of -5."
},
{
"role": "assistant",
"content": "sudo nice -n -5 ./import.sh"
}
] | single | nice |
[
{
"role": "system",
"content": "You are a Bash expert."
},
{
"role": "user",
"content": "How do I count how many cron processes are running?"
},
{
"role": "assistant",
"content": "pgrep -c cron"
}
] | pipeline | pgrep |
[
{
"role": "system",
"content": "You are a helpful assistant that writes correct, idiomatic Bash commands."
},
{
"role": "user",
"content": "What is tomcat running right now?"
},
{
"role": "assistant",
"content": "w tomcat"
}
] | single | w |
[
{
"role": "system",
"content": "You are a helpful assistant that writes correct, idiomatic Bash commands."
},
{
"role": "user",
"content": "Show the process tree including command-line arguments."
},
{
"role": "assistant",
"content": "pstree -a"
}
] | single | pstree |
[
{
"role": "system",
"content": "You are a Bash expert."
},
{
"role": "user",
"content": "Show established TCP connections numerically."
},
{
"role": "assistant",
"content": "netstat -tn"
}
] | single | netstat |
[
{
"role": "system",
"content": "You translate natural-language requests into correct Bash commands."
},
{
"role": "user",
"content": "Bump process 6994 to a higher priority (nice -5). Requires root."
},
{
"role": "assistant",
"content": "sudo renice -n -5 -p 6994"
}
] | single | renice |
[
{
"role": "system",
"content": "You are a Linux shell expert. Respond with Bash only."
},
{
"role": "user",
"content": "Print ec2-user's process hierarchy including PIDs."
},
{
"role": "assistant",
"content": "pstree -p ec2-user"
}
] | single | pstree |
[
{
"role": "system",
"content": "You are a helpful assistant that writes correct, idiomatic Bash commands."
},
{
"role": "user",
"content": "Set changelog.sql to permissions 644."
},
{
"role": "assistant",
"content": "chmod 644 changelog.sql"
}
] | single | chmod |
[
{
"role": "system",
"content": "You are a Linux shell expert. Respond with Bash only."
},
{
"role": "user",
"content": "I need getopts handling for three options: -i, -d and -t; print usage on bad input."
},
{
"role": "assistant",
"content": "#!/usr/bin/env bash\nset -euo pipefail\ni... | script | getopts |
[
{
"role": "system",
"content": "You translate natural-language requests into correct Bash commands."
},
{
"role": "user",
"content": "Print current vmstat counters once."
},
{
"role": "assistant",
"content": "vmstat"
}
] | single | vmstat |
[
{
"role": "system",
"content": "You are a Bash expert."
},
{
"role": "user",
"content": "Display kernel messages from the journal with priority error or worse."
},
{
"role": "assistant",
"content": "journalctl -k -p err"
}
] | single | journalctl |
[
{
"role": "system",
"content": "You are a helpful assistant that writes correct, idiomatic Bash commands."
},
{
"role": "user",
"content": "Download https://db01.example.com/file.zip and save it with its remote name."
},
{
"role": "assistant",
"content": "curl -O https://db01.example... | single | curl |
[
{
"role": "system",
"content": "You are a helpful assistant that writes correct, idiomatic Bash commands."
},
{
"role": "user",
"content": "Bump process 6456 to a higher priority (nice -5). Requires root."
},
{
"role": "assistant",
"content": "sudo renice -n -5 -p 6456"
}
] | single | renice |
[
{
"role": "system",
"content": "You are a Bash expert."
},
{
"role": "user",
"content": "Lower the priority of process 1145 to nice value 10."
},
{
"role": "assistant",
"content": "renice -n 10 -p 1145"
}
] | single | renice |
[
{
"role": "system",
"content": "You are a helpful assistant that writes correct, idiomatic Bash commands."
},
{
"role": "user",
"content": "Show TCP connections to api.internal."
},
{
"role": "assistant",
"content": "ss -tn dst api.internal"
}
] | single | ss |
[
{
"role": "system",
"content": "You are a command-line assistant specializing in Bash and GNU coreutils."
},
{
"role": "user",
"content": "Log vmstat output to a file with timestamps every 5 seconds."
},
{
"role": "assistant",
"content": "vmstat 5 | while read line; do echo \"$(date ... | pipeline | vmstat |
[
{
"role": "system",
"content": "You are a Bash expert."
},
{
"role": "user",
"content": "Create a script that dumps every MySQL database into its own gzip file under /mnt/backup, keeping 14 days."
},
{
"role": "assistant",
"content": "#!/usr/bin/env bash\nset -euo pipefail\ndest=/mnt... | script | mysqldump |
[
{
"role": "system",
"content": "You are a Linux shell expert. Respond with Bash only."
},
{
"role": "user",
"content": "Launch 'dd if=/dev/zero of=test.img bs=1M count=1024' at reduced priority (1)."
},
{
"role": "assistant",
"content": "nice -n 1 dd if=/dev/zero of=test.img bs=1M co... | single | nice |
[
{
"role": "system",
"content": "You are a Bash expert."
},
{
"role": "user",
"content": "Show vmstat memory counters in megabytes."
},
{
"role": "assistant",
"content": "vmstat -s -S M"
}
] | single | vmstat |
[
{
"role": "system",
"content": "You are a Bash expert."
},
{
"role": "user",
"content": "Report disk I/O counters via vmstat -d."
},
{
"role": "assistant",
"content": "vmstat -d"
}
] | single | vmstat |
[
{
"role": "system",
"content": "You translate natural-language requests into correct Bash commands."
},
{
"role": "user",
"content": "Show memory usage with a total line for RAM plus swap."
},
{
"role": "assistant",
"content": "free -h -t"
}
] | single | free |
[
{
"role": "system",
"content": "You are a Bash expert."
},
{
"role": "user",
"content": "monitor system performance with vmstat every 5 seconds using human-readable units"
},
{
"role": "assistant",
"content": "vmstat -S M 5"
}
] | single | vmstat |
[
{
"role": "system",
"content": "You are a command-line assistant specializing in Bash and GNU coreutils."
},
{
"role": "user",
"content": "List all files opened by cron."
},
{
"role": "assistant",
"content": "lsof -c cron"
}
] | single | lsof |
[
{
"role": "system",
"content": "You are a helpful assistant that writes correct, idiomatic Bash commands."
},
{
"role": "user",
"content": "Show the process tree sorted numerically by PID."
},
{
"role": "assistant",
"content": "pstree -n"
}
] | single | pstree |
[
{
"role": "system",
"content": "You translate natural-language requests into correct Bash commands."
},
{
"role": "user",
"content": "When was the system booted, per who -b?"
},
{
"role": "assistant",
"content": "who -b"
}
] | single | uptime |
[
{
"role": "system",
"content": "You are a Linux shell expert. Respond with Bash only."
},
{
"role": "user",
"content": "Write a countdown script from 10 to 1 that prints each number one second apart."
},
{
"role": "assistant",
"content": "#!/usr/bin/env bash\nset -euo pipefail\ni=10\... | script | while |
[
{
"role": "system",
"content": "You translate natural-language requests into correct Bash commands."
},
{
"role": "user",
"content": "Show a short w listing without the login/JCPU/PCPU columns."
},
{
"role": "assistant",
"content": "w -s"
}
] | single | w |
[
{
"role": "system",
"content": "You are a Bash expert."
},
{
"role": "user",
"content": "Show logged-in users with their IP addresses instead of hostnames."
},
{
"role": "assistant",
"content": "w -i"
}
] | single | w |
[
{
"role": "system",
"content": "You translate natural-language requests into correct Bash commands."
},
{
"role": "user",
"content": "Show detailed per-device iostat, 3s apart, 10 times."
},
{
"role": "assistant",
"content": "iostat -x 3 10"
}
] | single | iostat |
[
{
"role": "system",
"content": "You are a Linux shell expert. Respond with Bash only."
},
{
"role": "user",
"content": "Watch the kernel log live."
},
{
"role": "assistant",
"content": "dmesg -w"
}
] | single | dmesg |
[
{
"role": "system",
"content": "You translate natural-language requests into correct Bash commands."
},
{
"role": "user",
"content": "Print just the short hostname."
},
{
"role": "assistant",
"content": "hostname -s"
}
] | single | hostname |
[
{
"role": "system",
"content": "You are a command-line assistant specializing in Bash and GNU coreutils."
},
{
"role": "user",
"content": "For event.sql, show the last 20 lines."
},
{
"role": "assistant",
"content": "tail -n 20 event.sql"
}
] | single | tail |
[
{
"role": "system",
"content": "You are a Linux shell expert. Respond with Bash only."
},
{
"role": "user",
"content": "strip leading whitespace from every line of config.json and save to clean.txt"
},
{
"role": "assistant",
"content": "sed 's/^[[:space:]]*//' config.json > clean.txt... | pipeline | sed |
[
{
"role": "system",
"content": "You are a helpful assistant that writes correct, idiomatic Bash commands."
},
{
"role": "user",
"content": "Show all TCP sockets numerically."
},
{
"role": "assistant",
"content": "ss -tan"
}
] | single | ss |
[
{
"role": "system",
"content": "You are a helpful assistant that writes correct, idiomatic Bash commands."
},
{
"role": "user",
"content": "Fetch JSON from https://registry.example.org/api/users and pretty-print it with jq."
},
{
"role": "assistant",
"content": "curl -s https://regis... | pipeline | curl |
[
{
"role": "system",
"content": "You are a Linux shell expert. Respond with Bash only."
},
{
"role": "user",
"content": "Get the bare PID that holds port 443."
},
{
"role": "assistant",
"content": "lsof -i :443 -t"
}
] | single | lsof |
[
{
"role": "system",
"content": "You are a Linux shell expert. Respond with Bash only."
},
{
"role": "user",
"content": "strip leading whitespace from every line of settings.ini and save to clean.txt"
},
{
"role": "assistant",
"content": "sed 's/^[[:space:]]*//' settings.ini > clean.t... | pipeline | sed |
[
{
"role": "system",
"content": "You translate natural-language requests into correct Bash commands."
},
{
"role": "user",
"content": "Print how many CPU cores this machine has."
},
{
"role": "assistant",
"content": "echo \"Cores: $(nproc)\""
}
] | pipeline | cmdsub |
[
{
"role": "system",
"content": "You are a Linux shell expert. Respond with Bash only."
},
{
"role": "user",
"content": "On this box, show which processes are listening and on which ports."
},
{
"role": "assistant",
"content": "ss -tlnp | awk 'NR>1 {print $4, $6}'"
}
] | pipeline | ss |
[
{
"role": "system",
"content": "You translate natural-language requests into correct Bash commands."
},
{
"role": "user",
"content": "Extract the idle-CPU column from a vmstat snapshot."
},
{
"role": "assistant",
"content": "vmstat 1 2 | tail -n1 | awk '{print $15}'"
}
] | pipeline | vmstat |
[
{
"role": "system",
"content": "You are a helpful assistant that writes correct, idiomatic Bash commands."
},
{
"role": "user",
"content": "Launch 'make -j4' at reduced priority (5)."
},
{
"role": "assistant",
"content": "nice -n 5 make -j4"
}
] | single | nice |
[
{
"role": "system",
"content": "You are a Linux shell expert. Respond with Bash only."
},
{
"role": "user",
"content": "Show a summary of socket counts."
},
{
"role": "assistant",
"content": "ss -s"
}
] | single | ss |
[
{
"role": "system",
"content": "You are a Bash expert."
},
{
"role": "user",
"content": "Find files in include larger than +100M."
},
{
"role": "assistant",
"content": "find include -type f -size +100M"
}
] | single | find |
[
{
"role": "system",
"content": "You are a Linux shell expert. Respond with Bash only."
},
{
"role": "user",
"content": "Write a Bash script that alerts if the root filesystem is more than 85% full."
},
{
"role": "assistant",
"content": "#!/usr/bin/env bash\nset -euo pipefail\nthresho... | script | if |
[
{
"role": "system",
"content": "You translate natural-language requests into correct Bash commands."
},
{
"role": "user",
"content": "How many active sessions are there per w?"
},
{
"role": "assistant",
"content": "w | wc -l"
}
] | single | w |
[
{
"role": "system",
"content": "You are a helpful assistant that writes correct, idiomatic Bash commands."
},
{
"role": "user",
"content": "Show TCP sockets in the TIME-WAIT state with their process."
},
{
"role": "assistant",
"content": "ss -tp state time-wait"
}
] | single | ss |
[
{
"role": "system",
"content": "You are a Bash expert."
},
{
"role": "user",
"content": "Show kernel ring buffer messages with human-readable timestamps."
},
{
"role": "assistant",
"content": "dmesg -T"
}
] | single | dmesg |
[
{
"role": "system",
"content": "You are a Bash expert."
},
{
"role": "user",
"content": "Show listening TCP/UDP sockets and their PIDs."
},
{
"role": "assistant",
"content": "netstat -tulpn"
}
] | single | netstat |
[
{
"role": "system",
"content": "You are a command-line assistant specializing in Bash and GNU coreutils."
},
{
"role": "user",
"content": "List connections stuck in TIME-WAIT."
},
{
"role": "assistant",
"content": "ss -tp state time-wait"
}
] | single | ss |
[
{
"role": "system",
"content": "You are a helpful assistant that writes correct, idiomatic Bash commands."
},
{
"role": "user",
"content": "How long has the system been running?"
},
{
"role": "assistant",
"content": "uptime"
}
] | single | uptime |
[
{
"role": "system",
"content": "You are a Linux shell expert. Respond with Bash only."
},
{
"role": "user",
"content": "Add a timestamp column to vmstat, 3s interval, 10 rounds."
},
{
"role": "assistant",
"content": "vmstat -t 3 10"
}
] | single | vmstat |
[
{
"role": "system",
"content": "You are a helpful assistant that writes correct, idiomatic Bash commands."
},
{
"role": "user",
"content": "Check what holds port 5672 with ss."
},
{
"role": "assistant",
"content": "ss -tlnp sport = :5672"
}
] | single | ss |
[
{
"role": "system",
"content": "You are a Linux shell expert. Respond with Bash only."
},
{
"role": "user",
"content": "Count how many established TCP connections there are."
},
{
"role": "assistant",
"content": "ss -t state established | tail -n +2 | wc -l"
}
] | pipeline | ss |
[
{
"role": "system",
"content": "You are a Linux shell expert. Respond with Bash only."
},
{
"role": "user",
"content": "Print just the PID using port 8000."
},
{
"role": "assistant",
"content": "lsof -i :8000 -t"
}
] | single | lsof |
[
{
"role": "system",
"content": "You are a Linux shell expert. Respond with Bash only."
},
{
"role": "user",
"content": "Show the kernel routing table."
},
{
"role": "assistant",
"content": "netstat -rn"
}
] | single | netstat |
[
{
"role": "system",
"content": "You are a Bash expert."
},
{
"role": "user",
"content": "List the unique 'critical' messages in messages."
},
{
"role": "assistant",
"content": "grep 'critical' messages | sort -u"
}
] | pipeline | grep |
[
{
"role": "system",
"content": "You are a command-line assistant specializing in Bash and GNU coreutils."
},
{
"role": "user",
"content": "Show the process tree with PIDs, highlighting the current process."
},
{
"role": "assistant",
"content": "pstree -hp"
}
] | single | pstree |
[
{
"role": "system",
"content": "You are a Bash expert."
},
{
"role": "user",
"content": "Check if results.html exists and cat it, otherwise say it's missing."
},
{
"role": "assistant",
"content": "[ -f results.html ] && cat results.html || echo 'not found'"
}
] | pipeline | chain |
[
{
"role": "system",
"content": "You are a command-line assistant specializing in Bash and GNU coreutils."
},
{
"role": "user",
"content": "List all files opened by redis-server."
},
{
"role": "assistant",
"content": "lsof -c redis-server"
}
] | single | lsof |
[
{
"role": "system",
"content": "You are a Bash expert."
},
{
"role": "user",
"content": "Show only the load averages."
},
{
"role": "assistant",
"content": "uptime | awk -F'load average:' '{print $2}'"
}
] | single | uptime |
[
{
"role": "system",
"content": "You are a Bash expert."
},
{
"role": "user",
"content": "Download https://8.8.8.8/file.zip and save it with its remote name."
},
{
"role": "assistant",
"content": "curl -O https://8.8.8.8/file.zip"
}
] | single | curl |
[
{
"role": "system",
"content": "You are a Bash expert."
},
{
"role": "user",
"content": "List all listening TCP and UDP sockets with programs."
},
{
"role": "assistant",
"content": "ss -tulnp"
}
] | single | ss |
[
{
"role": "system",
"content": "You are a Bash expert."
},
{
"role": "user",
"content": "Timestamped vmstat every 3s for 5 samples."
},
{
"role": "assistant",
"content": "vmstat -t 3 5"
}
] | single | vmstat |
[
{
"role": "system",
"content": "You are a Bash expert."
},
{
"role": "user",
"content": "List active TCP sessions with numeric addresses."
},
{
"role": "assistant",
"content": "netstat -tn"
}
] | single | netstat |
[
{
"role": "system",
"content": "You are a Linux shell expert. Respond with Bash only."
},
{
"role": "user",
"content": "Print just the PID using port 80."
},
{
"role": "assistant",
"content": "lsof -i :80 -t"
}
] | single | lsof |
[
{
"role": "system",
"content": "You are a helpful assistant that writes correct, idiomatic Bash commands."
},
{
"role": "user",
"content": "Show a one-shot CPU and disk I/O summary."
},
{
"role": "assistant",
"content": "iostat"
}
] | single | iostat |
[
{
"role": "system",
"content": "You are a Bash expert."
},
{
"role": "user",
"content": "Print git's process hierarchy including PIDs."
},
{
"role": "assistant",
"content": "pstree -p git"
}
] | single | pstree |
[
{
"role": "system",
"content": "You are a helpful assistant that writes correct, idiomatic Bash commands."
},
{
"role": "user",
"content": "Display the process hierarchy for user deploy."
},
{
"role": "assistant",
"content": "pstree deploy"
}
] | single | pstree |
[
{
"role": "system",
"content": "You translate natural-language requests into correct Bash commands."
},
{
"role": "user",
"content": "Give 'find / -name '*.core'' a boosted CPU priority of -5."
},
{
"role": "assistant",
"content": "sudo nice -n -5 find / -name '*.core'"
}
] | single | nice |
[
{
"role": "system",
"content": "You are a Bash expert."
},
{
"role": "user",
"content": "Show TCP connections to db01.example.org."
},
{
"role": "assistant",
"content": "ss -tn dst db01.example.org"
}
] | single | ss |
[
{
"role": "system",
"content": "You translate natural-language requests into correct Bash commands."
},
{
"role": "user",
"content": "Launch 'ffmpeg -i in.mp4 out.mp4' at reduced priority (15)."
},
{
"role": "assistant",
"content": "nice -n 15 ffmpeg -i in.mp4 out.mp4"
}
] | single | nice |
[
{
"role": "system",
"content": "You are a Bash expert."
},
{
"role": "user",
"content": "Show the 50 most recent grafana-server log lines, newest last."
},
{
"role": "assistant",
"content": "journalctl -u grafana-server -n 50"
}
] | single | journalctl |
[
{
"role": "system",
"content": "You are a Linux shell expert. Respond with Bash only."
},
{
"role": "user",
"content": "Print the used percentage of the root filesystem as a number."
},
{
"role": "assistant",
"content": "df -h / | awk 'NR==2 {print $5}'"
}
] | pipeline | df |
[
{
"role": "system",
"content": "You are a helpful assistant that writes correct, idiomatic Bash commands."
},
{
"role": "user",
"content": "download https://db01.example.com/list.txt and count the lines"
},
{
"role": "assistant",
"content": "curl -s https://db01.example.com/list.txt ... | pipeline | curl |
[
{
"role": "system",
"content": "You are a Linux shell expert. Respond with Bash only."
},
{
"role": "user",
"content": "list only the .gif files in lib"
},
{
"role": "assistant",
"content": "ls lib | grep '\\.gif$'"
}
] | pipeline | ls |
[
{
"role": "system",
"content": "You are a Linux shell expert. Respond with Bash only."
},
{
"role": "user",
"content": "Include a combined RAM+swap total in the free output."
},
{
"role": "assistant",
"content": "free -h -t"
}
] | single | free |
[
{
"role": "system",
"content": "You are a command-line assistant specializing in Bash and GNU coreutils."
},
{
"role": "user",
"content": "List active (established) sessions on port 53."
},
{
"role": "assistant",
"content": "lsof -i :53 -sTCP:ESTABLISHED"
}
] | single | lsof |
[
{
"role": "system",
"content": "You are a Linux shell expert. Respond with Bash only."
},
{
"role": "user",
"content": "Show detailed per-device iostat, 2s apart, 3 times."
},
{
"role": "assistant",
"content": "iostat -x 2 3"
}
] | single | iostat |
[
{
"role": "system",
"content": "You are a helpful assistant that writes correct, idiomatic Bash commands."
},
{
"role": "user",
"content": "Tally how many times each log level appears in audit.log."
},
{
"role": "assistant",
"content": "grep -oE '(INFO|WARN|ERROR|DEBUG)' audit.log | ... | pipeline | grep |
[
{
"role": "system",
"content": "You translate natural-language requests into correct Bash commands."
},
{
"role": "user",
"content": "Show the process tree for user dave."
},
{
"role": "assistant",
"content": "pstree dave"
}
] | single | pstree |
[
{
"role": "system",
"content": "You are a Linux shell expert. Respond with Bash only."
},
{
"role": "user",
"content": "Sum the response sizes (column 10) in metrics.log."
},
{
"role": "assistant",
"content": "awk '{s+=$10} END{print s}' metrics.log"
}
] | pipeline | awk |
[
{
"role": "system",
"content": "You are a Linux shell expert. Respond with Bash only."
},
{
"role": "user",
"content": "Run 'rsync -a /data /mnt/backup' at a lower priority (nice 19)."
},
{
"role": "assistant",
"content": "nice -n 19 rsync -a /data /mnt/backup"
}
] | single | nice |
[
{
"role": "system",
"content": "You are a Bash expert."
},
{
"role": "user",
"content": "Show memory usage using powers of 1000 (SI units)."
},
{
"role": "assistant",
"content": "free -h --si"
}
] | single | free |
[
{
"role": "system",
"content": "You translate natural-language requests into correct Bash commands."
},
{
"role": "user",
"content": "Show the systemd journal for the apache2 service."
},
{
"role": "assistant",
"content": "journalctl -u apache2"
}
] | single | journalctl |
[
{
"role": "system",
"content": "You are a command-line assistant specializing in Bash and GNU coreutils."
},
{
"role": "user",
"content": "Show every network socket in use."
},
{
"role": "assistant",
"content": "lsof -i"
}
] | single | lsof |
[
{
"role": "system",
"content": "You translate natural-language requests into correct Bash commands."
},
{
"role": "user",
"content": "Show just the amount of free memory in MB."
},
{
"role": "assistant",
"content": "free -m | awk '/Mem/ {print $4}'"
}
] | pipeline | free |
End of preview. Expand in Data Studio
Bash Instruct I — 55,000 verified natural-language → Bash pairs
⚠️ Superseded by Bash Instruct III
III doubles the utility vocabulary (89 → 182), adds grouped equivalent answers, real human phrasing from tldr-pages, validation by execution on real Linux, and a style pass that removes shell antipatterns. New work should use III.
This version remains useful for one specific purpose: it overlaps III by only ~29%, so it is the one generation that can be deduplicated in as extra volume. See Combining versions.
Bash Instruct I is the first-generation synthetic instruction-tuning dataset in this family, pairing natural-language requests with correct Bash: single commands, short pipelines, and multi-line scripts. It is built for supervised fine-tuning of small LLMs that must turn a plain request into shell code that actually runs.
Every row is a three-turn chat conversation (system / user / assistant) plus two
metadata fields (category, utility) for slicing and analysis.
| 🤗 Dataset viewer | https://huggingface.co/datasets/Frost2o24/bash-instruct-55k |
| 🐙 Source, generator & validator | https://github.com/ya5h-P/bash-instruct-55k |
| 📦 Later generations | III (recommended) · II |
| ⚖️ License | MIT |
from datasets import load_dataset
ds = load_dataset("Frost2o24/bash-instruct-55k", split="train")
Version comparison
| I (this) | II | III | |
|---|---|---|---|
| Rows | 55,000 | 54,803 | 54,360 |
| Distinct primary utilities | 89 | 182 | 182 |
variant_group (grouped equivalent answers) |
✗ | ✓ | ✓ |
| Real phrasing seeded from tldr-pages | ✗ | ✓ | ✓ |
| Validated by execution on real Linux | ✗ | ✓ | ✓ |
| ShellCheck-clean rate (warning level) | 97.9% | 98.5% | 99.8% |
Antipattern style pass (SC2002, SC2010) |
✗ | ✗ | ✓ |
This version is validated statically only — bash -n, ShellCheck, and the
argument-fidelity gate. It was never executed at scale on real Linux, and it has a narrower
utility vocabulary. Prefer III.
At a glance
All figures below are recomputed directly from the shipped bash_dataset.jsonl, not
carried over from an earlier run.
| Metric | Value |
|---|---|
| Rows | 55,000 |
single / pipeline / script |
40% / 35% / 25% (exact) |
| Distinct primary utilities | 89 |
bash -n syntax-valid |
100% (0 / 55,000 failures) |
| ShellCheck-clean at warning level | 97.9% of a 6,000-row random sample |
Exact-duplicate (request, command) pairs |
0 |
| Unique request text | 98.7% |
| Unique command text | 75.6% |
| Argument fidelity | 100% (0 request/command noun mismatches) |
| Max share of any single utility | 4.0% (hard cap of 2,200 rows per utility) |
| Most common 4-word request opening | 1.4% of rows |
| Mean command length | 85 chars · 13.8 tokens |
Note the trade-off against later versions: request text here is more unique (98.7% vs ~80% of rows) because there are no grouped answer variants, but command text is less unique (75.6% vs 78.7%) because the same command recurs under different phrasings.
Quickstart
Load
from datasets import load_dataset
ds = load_dataset("Frost2o24/bash-instruct-55k", split="train")
print(ds)
# Dataset({features: ['messages', 'category', 'utility'], num_rows: 55000})
print(ds[0]["messages"])
Fine-tune (TRL SFTTrainer)
The messages column is already in the OpenAI-style chat format that TRL, Axolotl,
Unsloth, and LLaMA-Factory consume directly — no preprocessing or column mapping needed.
from datasets import load_dataset
from trl import SFTTrainer, SFTConfig
ds = load_dataset("Frost2o24/bash-instruct-55k", split="train")
trainer = SFTTrainer(
model="Qwen/Qwen2.5-Coder-1.5B-Instruct",
train_dataset=ds, # `messages` is auto-detected
args=SFTConfig(
output_dir="bash-sft",
max_length=1024,
num_train_epochs=2,
per_device_train_batch_size=8,
),
)
trainer.train()
Analyze and slice
import collections
print(collections.Counter(ds["category"]))
print(collections.Counter(ds["utility"]).most_common(15))
scripts = ds.filter(lambda r: r["category"] == "script")
There is no variant_group in this version, so a plain random row split is safe — every
row is a distinct (request, command) pair.
Inspect without Python
head -n 1 bash_dataset.jsonl | jq .
jq -r '.utility' bash_dataset.jsonl | sort | uniq -c | sort -rn | head -20
jq -r '.category' bash_dataset.jsonl | sort | uniq -c
Dataset structure
One JSON object per line in bash_dataset.jsonl:
{
"messages": [
{"role": "system", "content": "You are a Bash expert."},
{"role": "user", "content": "Show the last 20 lines of error.log."},
{"role": "assistant", "content": "tail -n 20 error.log"}
],
"category": "single",
"utility": "tail"
}
| Field | Type | Description |
|---|---|---|
messages |
list[{role, content}] |
The training conversation, always exactly system → user → assistant. |
category |
string |
single · pipeline · script. See below. |
utility |
string |
Primary command of the solution (grep, awk, find, for, …). Used to enforce the per-command cap, and useful for slicing or re-balancing training. |
Verified schema integrity: all 55,000 lines parse as JSON, all carry the same three
keys, all have exactly the system/user/assistant role order, and no message content
is empty.
Categories
category |
Rows | Share | What it contains |
|---|---|---|---|
single |
22,000 | 40% | One command, no pipe or chaining. tail -n 20 error.log |
pipeline |
19,250 | 35% | Pipes, &&/||, $(...), xargs. ps aux | awk '$3>50 {print $2}' |
script |
13,750 | 25% | Multi-line Bash (JSON-escaped \n), mean 9.8 lines. |
What's in the scripts
| Feature | Share of script rows |
|---|---|
set -euo pipefail header |
100% |
for loop |
47.8% |
if [ … ] test |
13.8% |
while loop |
11.9% |
getopts argument parsing |
9.7% |
trap cleanup |
0.1% |
Here-docs (<<) |
0.1% |
Scripts are short, safe-by-default operational snippets — not large programs. trap
and here-doc usage is deliberately rare, and there are no function definitions.
System & info utility coverage
Small models routinely fumble the observability and process-management tools. These carry enforced minimum counts with correct, idiomatic flag usage:
| Utility | Rows | Utility | Rows | Utility | Rows |
|---|---|---|---|---|---|
journalctl |
2,200 | du |
806 | pstree |
782 |
lsof |
673 | ss |
332 | renice |
296 |
nice |
293 | vmstat |
286 | ps |
242 |
netstat |
239 | iostat |
184 | w |
158 |
dmesg |
156 | systemctl |
111 | free |
90 |
df |
45 | uptime |
34 | hostname |
25 |
uptime and hostname are lower because their genuine idiomatic command space is small;
they are covered with real flag variants rather than padded with synonymous phrasings.
System prompts
Five equivalent Bash-assistant system prompts are rotated near-uniformly (~11.0k rows each) so the model does not overfit one string.
Quality assurance
1. Syntax — bash -n on every command
0 failures out of 55,000 (100% pass), verified with GNU bash 5.2.37 on Debian.
2. Static analysis — ShellCheck at warning level
On a 6,000-row random sample (ShellCheck 0.10.0): 97.95% of commands are completely clean. 123 rows produced a finding, almost all of them a single style nit:
| Code | Findings | What it is |
|---|---|---|
SC2010 |
115 | ls | grep instead of a glob — 93% of all findings |
SC2046 |
3 | Unquoted command substitution |
SC2062 |
3 | Unquoted grep pattern |
SC2164 |
2 | cd without || exit |
All findings are warning level; there are no ShellCheck errors. The SC2010 pattern is
fixed at the generator level in
Bash Instruct III.
3. Argument fidelity
Every pair is gated so the command actually references the concrete nouns the request names — the same filename, extension, user, group, service, process, and port. The shipped file has 0 mismatches.
This gate exists because of a real failure in an earlier build of this generator: roughly
12% of pairs had the right command shape but the wrong concrete noun — a different
filename than the request asked for. Neither bash -n nor ShellCheck can detect that,
because a wrong filename still parses and still lints clean. argcheck.py closes exactly
that gap, and it is the most important quality lesson in this dataset family.
Not validated by execution
Unlike II and
III, this version was
never executed at scale on real Linux. Many commands (systemctl, journalctl, apt,
vmstat) are idiomatic but were validated statically only. The whole-set execution pass
applied to II found 43 commands that were malformed in ways static analysis missed;
comparable defects are likely present here and unremoved.
Reproduce the audit yourself
# argument-fidelity audit over the shipped file
python argcheck.py --data bash_dataset.jsonl
# re-run bash -n over the whole file
python generate.py validate-all
The standalone validate.py — including the sandboxed execution modes — ships with
II and
III and can be pointed
at this file.
Provenance
How it is built
Generated by a recipe engine (generate.py, included and reproducible). Each recipe family
emits (request, command) pairs by combining hand-written phrasing templates
(imperative / question / casual) with realistic parameter pools: plausible filenames,
directories, ports, services, users, patterns.
Every parameter is drawn once per example and reused in both the request and the command, so the two can never disagree.
The generator enforces:
- category quotas (40 / 35 / 25),
- a 4% per-utility cap (2,200 rows) so no command dominates — a real failure mode of earlier runs,
- minimum coverage floors for the system/info utilities above,
- SHA-1 deduplication of
(request, command)pairs, - an argument-fidelity gate (
argcheck.py) — the command must reference the same concrete nouns the request names, or the pair is rejected, - phrasing diversification so no single opening template dominates,
- a
bash -nsyntax gate on every command before it is written.
Regenerating from source
python generate.py plan # quota table + registered recipes
python generate.py run # generate to ./bash_dataset.jsonl (resumable)
python generate.py status # where a run stands
python generate.py validate-all # re-run bash -n over the whole file
bash_dataset.jsonl is ground truth and a derived progress.json snapshot is rewritten
after every batch, so an interrupted run continues where it stopped.
Files
| File | Description |
|---|---|
bash_dataset.jsonl |
The dataset — 55,000 rows, ~20 MB (no Git LFS needed). |
generate.py |
The generator. Reproducible and resumable. |
argcheck.py |
Argument-fidelity checker — generation gate and standalone auditor. |
generate.py and argcheck.py are pure Python 3 with no third-party dependencies.
Loading the dataset requires only datasets (or nothing at all — it is plain JSONL).
Combining versions
Overlap of (request, command) pairs, measured across the shipped files:
| Pair | Shared pairs | Share of the smaller set |
|---|---|---|
| I ∩ III | 15,896 | 29.2% — mostly distinct, safe to merge |
| I ∩ II | 18,200 | 33.2% |
| II ∩ III | 50,715 | 92.5% — never concatenate those two |
This is version I's main remaining use. Because it overlaps III by only ~29%, deduplicating the two together yields roughly 93.5k unique pairs — meaningfully more volume than either alone. Weight it below III, which is measurably higher quality.
from datasets import load_dataset, concatenate_datasets
a = load_dataset("Frost2o24/bash-instruct-III-55k", split="train")
b = load_dataset("Frost2o24/bash-instruct-55k", split="train")
b = b.add_column("variant_group", [f"v1-{i}" for i in range(len(b))])
merged, seen = concatenate_datasets([a, b]), set()
merged = merged.filter(
lambda r: not ((k := (r["messages"][1]["content"], r["messages"][2]["content"])) in seen
or seen.add(k))
)
print(len(merged)) # ~93.5k unique pairs
Intended uses & limitations
Intended uses
Teaching a small model to map natural-language requests to correct single Bash commands, short pipelines, and small scripts — including the system/info utilities listed above. In this dataset family its best role is as supplementary volume deduplicated alongside III, rather than as a primary training set.
Limitations
Please read these before reporting results.
- Superseded. III is better on every measured axis. Use I for extra volume, not as your primary corpus.
- Not executed. Static validation only; comparable to the 43 malformed commands that whole-set execution later found in II. Some are likely still present here.
- Synthetic. Variety comes from recombining templates and token pools, not from human authorship. Phrasing is diversified — 98.7% of request strings are unique and no 4-word opening exceeds 1.4% of rows — but the underlying request shapes come from a finite set of families, and ~24% of commands recur under different phrasings.
- Valid ≠ semantically correct.
bash -n, ShellCheck, and the argument gate prove that commands parse, lint, and use the nouns the request named. None of them prove the command's logic satisfies the intent. Some pairs are plausible-but-approximate — anawkcolumn index, for instance, assumes a particular log layout. There is no human correctness audit. - Single-turn, one canonical answer. No explanations, alternatives, negative examples, reasoning traces, or multi-turn dialogue.
- GNU/Linux-centric. Commands assume GNU coreutils and a systemd-based distribution.
BSD/macOS flag differences (
sed -i,stat,ps) are not covered, and neither arezsh/fish. - Short scripts only. Mean 9.8 lines, no function definitions, minimal
trapand here-doc coverage. Not a source for large structured shell programs. - Not safety-tuned. The corpus contains destructive-capable commands (
rm,chmod,chown,kill,dd) as legitimate answers to requests that ask for them. A model trained on it will emit such commands readily. Add refusal/confirmation behaviour separately if your deployment needs it, and never auto-execute model output.
Citation
If you use this dataset, please cite it:
@misc{pimplapure2026bashinstruct1,
title = {Bash Instruct: A Synthetic Instruction-Tuning Dataset for
Natural Language to Bash Generation},
author = {Pimplapure, Yash},
year = {2026},
publisher = {Hugging Face},
howpublished = {\url{https://huggingface.co/datasets/Frost2o24/bash-instruct-55k}},
note = {55,000 chat-formatted examples; 100\% \texttt{bash -n} valid,
97.9\% ShellCheck-clean. Superseded by Bash Instruct III}
}
License
MIT.
Supporting this work
This dataset — generation, whole-set Linux execution, and the fine-tuning runs used to check that it actually teaches the task — was built and validated on modest consumer hardware, which is the main thing limiting how far the next version can go.
If your team has an NVIDIA DGX Spark or an AMD Ryzen AI Max+ ("Strix Halo") AI dev kit to spare, it would go directly into larger validated corpora, real fine-tuned baselines, and published eval numbers for this dataset family. Reach out via the dataset discussions tab. No obligation either way — the data stays MIT and freely available regardless.
- Downloads last month
- 82