⚡ Pipes

Clean Email Strip formatting from pasted email for clean plain text Clean AI Output Clean plain text from ChatGPT or AI output Decode & Format JWT Decode a JWT and pretty-print header and payload Word Frequency Count word frequency in text Extract Emails Extract email addresses from text Fit Context Window Clean HTML and truncate to fit an LLM token budget Find Unique Errors Find unique error lines in logs JSON → Table Readable ASCII table from a JSON array Extract AI Code Numbered code blocks from AI responses Count Markdown Words Count words in Markdown content Clean HTML to Text Readable plain text from raw HTML Flatten & Sort JSON Sorted dot-notation view of nested JSON Transliterate to Slug Convert Unicode text to ASCII and create a URL slug Query & Format JSON Formatted JSON value from a nested path HTML Word Frequency Count word frequency in HTML content Extract & Sort Numbers Sorted list of all numbers found in text Split, Sort & Deduplicate Clean sorted list from comma-separated values Lowercase, Sort & Deduplicate Case-normalized sorted unique list Strip Comment Lines Clean text with comment lines removed HTML to Plain Text Plain text from HTML-encoded content Decode Base64 to JSON Readable JSON from Base64-encoded data Minify & Encode JSON Base64-encoded minified JSON Extract Unique Links Extract unique URLs from text Extract URLs Extract URLs from text Extract Unique Email Addresses Extract unique email addresses from text Sort & Deduplicate Sorted unique list of lines Number Lines Sorted, numbered unique lines Trim & Clean Clean text with no extra whitespace Clean & Count Text Line and word counts for cleaned text Remove HTML Tags Plain text with HTML tags removed Reverse & Number Numbered lines in reverse order Reverse Sort & Number Numbered lines in reverse-sorted order Markdown Table GitHub Markdown table from CSV data Base64 Encode Encode text as Base64 URL Encode Encode text for safe use in URLs Character Stats Character, word, and line counts

⟐ Data

csv-to-json csv-to-md-table csv-to-table json-flatten json-format json-keys json-minify json-query json-to-csv json-to-toml json-to-yaml sql-format toml-to-json yaml-to-json

</> Web

extract-codeblocks html-to-text md-to-html mermaid-render mustache-render strip-markdown syntax-highlight

{ } Encode

base64-decode base64-encode escape html-decode html-encode jwt-decode qr-encode url-decode url-encode

⇔ Decode

chmod-calc cron-explain nato

# Stats

char-freq token-chunk token-count token-split token-truncate wc word-freq

.* Regex

regex-extract regex-replace

⚡ Gen

ascii-art hash lorem timestamp uuid

◷ Time

epoch-to-human human-to-epoch

Fit Context Window

Clean HTML and truncate to fit an LLM token budget

Analyze

Input

How it works

1. html-to-text Convert HTML to plain text preserving line breaks 2. squeeze Collapse extra whitespace and blank lines 3. token-truncate Truncate text to fit a token budget (real BPE)

$ html-to-text | squeeze | token-truncate --max=4096

Examples

Clean a documentation page to fit a 30-token LLM prompt

Usage

"<article><h2>Getting Started with React</h2><p>React is a Ja..." | html-to-text | squeeze | token-truncate --max=4096

Strip blog HTML and extra whitespace for an API summary call

Usage

"<div class="post"><span class="meta">Posted by admin &bull; ..." | html-to-text | squeeze | token-truncate --max=4096

Convert a status page table to clean text for an incident prompt

Usage

"<table><tr><th>Name</th><th>Status</th></tr><tr><td>API Gate..." | html-to-text | squeeze | token-truncate --max=4096

Combine with other tools