Brainstorm

Brainstorm is a focused text transformation tool designed to help generate and normalize candidate strings from raw text. It is particularly useful for tasks like wordlist creation, passphrase / token derivation, and transforming free-form text into structured candidate outputs.

Brainstorm is written in Go, is compatible with multiple platforms, and is designed to work well in Unix-style pipelines. It reads from standard input and writes transformed output to standard output.

Brainstorm is intentionally minimal. It focuses on:

• N‑gram generation from sentences.
• Filtering noisy or non-word-like input.
• Generating normalized, case-adjusted tokens with configurable length ranges.

Features

• Streaming stdin pipeline: Reads from standard input and writes to standard output, making it easy to chain with other tools.
• N‑gram Generation: Generates n‑grams over a configurable word-length range.
• Normalization & Cleanup:
  • Removes leading/trailing non-letter characters on each line.
  • Filters out lines that are unlikely to contain meaningful words.
  • Cleans common control and whitespace characters.
• Case Transformations:
  • Generates title-cased concatenations (e.g., "hello world" → "HelloWorld").
• Parallel Processing:
  • Uses a worker pool to process lines concurrently.

Install

From source with go:

go install github.com/hashcracky/brainstorm@latest

From git clone then build with go:

git clone https://github.com/hashcracky/brainstorm
cd brainstorm
go build ./main.go
mv ./main ~/go/bin/brainstorm
brainstorm

---

Basic Usage

Brainstorm reads from standard input and writes to standard output. There are no positional arguments; all behavior is controlled via flags.

Core Flags

• -w string
  • N‑gram word-length range in the form start-end.
  • Default: 1-5
• -l string
  • Final output length range in the form min-max.
  • Default: 4-32

Example:

cat source.txt | brainstorm -w 1-4 -l 6-20 > candidates.txt

Full Flags

Usage of Brainstorm version (0.1.4):

input | brainstorm [options] > output

Accepts standard input and writes transformed output to standard output.

Options:
  -l string
        Final output length range in the form min-max (for example, 4-32). (default "4-32")
  -unicode
        Include non-Latin multi-byte letter sequences by relaxing Latin vowel heuristics.
  -w string
        N-gram word length range in the form start-end (for example, 1-5). (default "1-5")