goldenseed
par beanapologist
Browser & AutomationRésumé
Deterministic entropy streams for reproducible testing and procedural generation.
Description
Ce que fait ce skill
GoldenSeed generates infinite deterministic byte streams from fixed seeds using the UniversalQKD class in the golden-seed Python package. Same seed always produces identical output across runs. Not cryptographically secure — suited for testing, procedural generation, and simulations where reproducibility matters more than unpredictability.
Python's built-in random module doesn't guarantee identical output across versions or platforms, so GoldenSeed fills that gap when exact byte-level reproducibility across runs and environments is required.
Quand l'utiliser
- Debugging flaky tests by replaying the exact random sequence that caused a failure
- Generating identical game worlds or map chunks from a shared numeric seed
- Running reproducible Monte Carlo simulations with verifiable byte-level results
- Creating procedural art or generative NFTs where the seed proves the output
- Proving fair dice rolls in competitive games by publishing the seed after the fact
Exemple de workflow
User asks: Generate reproducible test user records using seed 42
- 1
Install golden-seed via pip
- 2
Import UniversalQKD from gq and initialize a TestDataGenerator with seed=42
- 3
Call random_user() to get a deterministic dict with id, age, and premium flag
- 4
Re-initialize with seed=42 and call random_user() again
- 5
Assert both dicts are identical to confirm full reproducibility
Two identical user dicts confirming the sequence is byte-for-byte reproducible from seed 42
Installation
clawhub install beanapologist/goldenseedExécute la commande CLI, ou colle l'instruction dans Claude Code, Cursor, Codex ou l'agent de code que tu utilises.