vrk slug

Try it in your browser

vrk, compiled to WebAssembly

# Turns a title into a lowercase, hyphenated slug.

Command to run

echo 'My Blog Post: A Deep Dive into LLM Pipelines!' | vrk slug

Enter runs, Shift+Enter adds a line

output

Press RUN. Nothing leaves your browser.

files you can cat:

Edit the command and press RUN. Open the full playground for more recipes, or install vrk to run it on your own files.

vrk slug converts any text to a URL-safe, filename-safe slug with Unicode normalization.

The problem

A slugify function turns “Uber’s $3.1B Q4 Revenue” into “uber-s–3-1b-q4-revenue” with a double hyphen. Another title has Unicode characters that pass through unchanged and break a downstream system expecting ASCII-only paths. tr and sed handle simple cases but miss edge cases consistently.

The solution

vrk slug converts any text to a URL-safe, filename-safe slug. It folds Latin text to ASCII (Straße becomes strasse) and keeps other scripts, such as Devanagari and CJK, as they are. It lowercases the text, replaces non-alphanumeric characters with hyphens, and collapses consecutive hyphens. --max cuts the slug at a word boundary.

Before and after

Before

echo "$TITLE" | tr '[:upper:]' '[:lower:]' | tr ' ' '-' | tr -cd 'a-z0-9-'
# Misses Unicode normalization, doesn't collapse double hyphens

After

echo $TITLE | vrk slug

Example

echo 'My Blog Post: A Deep Dive into LLM Pipelines!' | vrk slug

Exit codes

Code	Meaning
0	Success
1	Runtime error (I/O failure)
2	Interactive TTY with no stdin

Flags

Flag	Short	Type	Description
`--separator`		string	Word separator character or string
`--max`		int	Max output length in characters; truncated at last separator (0 = no limit)
`--ascii`		bool	Keep only ASCII; other scripts are dropped (empty slug plus a warning if nothing is left)
`--json`	-j	bool	Emit JSON per line: {input, output}
`--quiet`	-q	bool	Suppress stderr output

How it works

vrk slug turns text into a URL-safe slug: lowercase words joined by hyphens.

$ echo 'Hello World' | vrk slug
hello-world

$ echo 'My Blog Post: A Deep Dive into LLM Pipelines!' | vrk slug
my-blog-post-a-deep-dive-into-llm-pipelines

Custom separator

$ echo 'Hello World' | vrk slug --separator _
hello_world

Length limit

--max cuts the slug at a word boundary, so a word is never cut in half:

$ echo 'My Very Long Blog Post Title That Goes On Forever' | vrk slug --max 30
my-very-long-blog-post-title

Batch processing

slug makes one slug per input line:

$ printf 'First Post\nSecond Post\nThird Post\n' | vrk slug
first-post
second-post
third-post

Non-Latin text

Latin letters fold to ASCII. slug keeps other scripts as they are, so a Hindi or Japanese title still gets a slug:

$ printf 'Straße\nनमस्ते दुनिया\n東京 2026\n' | vrk slug
strasse
नमस्ते-दुनिया
東京-2026

Every input line gives exactly one output line. If a line has nothing to keep, slug prints an empty line and warns on stderr, so the output lines match the input lines. Use --ascii when the target needs strict ASCII:

$ echo '東京 2026' | vrk slug --ascii
2026

JSON output

$ echo 'Hello World' | vrk slug --json
{"input":"Hello World","output":"hello-world"}

Pipeline integration

Generate cache keys from URLs

# Use slugified URLs as kv keys
for url in $(cat urls.txt); do
  KEY=$(echo "$url" | vrk slug)
  vrk kv set --ns cache "$KEY" "$(vrk grab "$url")" --ttl 24h
done

Create filenames from document titles

# Generate safe filenames for downloaded articles
while IFS= read -r url; do
  TITLE=$(vrk grab --json "$url" | jq -r '.title')
  FILENAME=$(echo "$TITLE" | vrk slug --max 50).md
  vrk grab "$url" > "$FILENAME"
done < urls.txt

When it fails

If you give no input on an interactive terminal, slug exits 2:

$ vrk slug
usage error: slug: no input
$ echo $?
2