CLI Reference

The parot binary is built from the crates/cli workspace member. Its full help tree is embedded below from _api/cli.md, regenerated by the api-docs skill against the live clap definitions.

For quickstart and common workflows, see the Guide.

---

CLI Reference: parot

Text index toolkit -- duplicate detection, pattern search, transforms, and more.

parot [COMMAND] [OPTIONS]

Backward compatibility: parot file.md is equivalent to parot scan file.md. Any first argument that is not a known subcommand and does not start with - is treated as a file path and scan is prepended automatically.

---

parot scan <file>

Find duplicate phrases in a text file.

parot scan <FILE> [OPTIONS]

Option	Type	Default	Description
--min-words	usize	4	Minimum words in a phrase
--min-chars	usize	9	Minimum characters excluding spaces
--max-words	usize	50	Maximum words in a phrase
--min-count	usize	2	Minimum occurrences to report
--top	usize	all	Show only top N results
--no-blocks	flag	off	Disable block detection
--json	flag	off	Output as JSON array
--quiet	flag	off	Suppress stats, only show results

Output (text): One line per duplicate: {count}x ({words} words) "{phrase}"

Output (JSON): Array of {"phrase": "...", "count": N, "words": N}

---

parot index-dump <file>

Print the raw index: one rank\tposition row per line.

parot index-dump <FILE> [OPTIONS]

Option	Type	Default	Description
--bytes	flag	off	Treat file as raw bytes instead of UTF-8 text
--json	flag	off	Output as JSON array

Output (text): Tab-separated rank\tposition per line.

---

parot prefix-stats <file>

Print adjacent shared-prefix lengths for the index.

parot prefix-stats <FILE> [OPTIONS]

Option	Type	Default	Description
--bytes	flag	off	Treat file as raw bytes instead of UTF-8 text
--json	flag	off	Output as JSON array

Output (text): Tab-separated rank\tvalue per line.

---

parot search <file> <pattern>

Search for a pattern and print match positions.

parot search <FILE> <PATTERN> [OPTIONS]

Option	Type	Default	Description
--bytes	flag	off	Treat file as raw bytes instead of UTF-8 text
--json	flag	off	Output as JSON array

Output (text): One byte position per line.

Output (JSON): Array of positions.

---

parot count <file> <pattern>

Count occurrences of a pattern.

parot count <FILE> <PATTERN> [OPTIONS]

Option	Type	Default	Description
--bytes	flag	off	Treat file as raw bytes instead of UTF-8 text

Output: A single integer count.

---

parot transform <file>

Print the reversible block transform of the input.

parot transform <FILE> [OPTIONS]

Option	Type	Default	Description
--bytes	flag	off	Treat file as raw bytes (output raw transformed bytes to stdout) instead of UTF-8 (output lossy UTF-8)

---

parot lcs <file1> <file2>

Find the longest common substring between two files.

parot lcs <FILE1> <FILE2>

No additional options. Both files are read as UTF-8 text.

Output: Length: N followed by the substring, or No common substring found. if none exists.

---

parot info <file>

Print file statistics: size, word count, line count, and distinct-substring count.

parot info <FILE> [OPTIONS]

Option	Type	Default	Description
--json	flag	off	Output as JSON object

Output (text):

Size:                 N bytes
Words:                N
Lines:                N
Distinct substrings:  N

Output (JSON): {"size_bytes": N, "words": N, "lines": N, "distinct_substrings": N}

---

Global Options

Option	Description
--help, -h	Print help
--version, -V	Print version