Destroy Integers

Reverse: deterministic – the digit order is flipped end-to-end. 1234 → 4321, 1000 → 1 (trailing zeros become leading zeros, then are stripped when re-parsed as an integer). Scramble: non-deterministic – uses Fisher-Yates shuffle to randomly permute the digits. 12345 → 53241 (or any of the 120 permutations). Sign preserved in both.

Differential privacy approximation, synthetic data generation, jittering data for testing. 100 with noise=10 → uniform random in [90, 110]. Each integer gets independently noised. The output preserves rough magnitude while obscuring exact values – useful for sharing data summaries where exact integers are sensitive. NOT cryptographically secure: the noise comes from Math.random() which is predictable; use a CSPRNG for real privacy work.

Random: replaces the integer with an UNRELATED random value in [min, max]. The original magnitude is lost. Use for full obfuscation or generating synthetic data.

Matching integers are entirely removed from the output. Output is shorter than input. With filter=all, delete mode produces empty output (everything matched). With filter=positive, only positives are removed. Useful for “drop the positives from this list” workflows. The stats card shows “Removed” count separately from “Destroyed” (Removed is the count of items that got dropped from the list; Destroyed is the count that underwent any operation).

After reversing, the result is re-parsed as an integer, so leading zeros are stripped. 1000 → 1, 1230 → 321, 10000 → 1. If you want to preserve digit count (output 00321 instead of 321), use the “Reverse Integer” tool’s preserve-leading-zeros option (sibling tool). This tool is integer-output by design.

Yes. The sign is preserved for Reverse and Scramble. For Random, the result is unrelated to input (uses min/max only). For Noise, sign preservation depends on magnitude: -5 with noise ±10 might become +3 or -15. For Delete with filter=negative, negatives are removed and positives stay. The filter selects which integers undergo the operation.

Not supported – uses Math.random(). For reproducible scrambling/random, you’d need to provide a seed and a seeded PRNG. That would add ~1 KB and complicate the API. If reproducibility is critical, run once and save the output, or use a separate seeded-PRNG tool. Most “destroy” use cases (privacy, obfuscation, synthetic data) want unpredictable randomness anyway.

NO. The output is recoverable in several ways: Random/Noise use predictable Math.random(). Reverse and Scramble are reversible if you know the original. Zero just replaces – original value irretrievable but anyone with the original data can re-derive. For actual cryptographic data destruction, use authenticated encryption with a discarded key, or hardware-level secure erase. This tool is for: obfuscating display data, generating test/sample integers, demonstrating digit transformations, anonymizing aggregated stats.

Yes for browser privacy (data never leaves your device). NOT for cryptographic destruction (see previous question). All processing happens in your browser via vanilla JavaScript.