Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
Updated docs: Emphasize testing reproducibility, perfect 50/50 coin flip, hash verification. Clear warnings about non-cryptographic use. Agent-focused examples for debugging flaky tests.
What This Skill Does
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.
When to use it
- 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
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: Generate reproducible test user records using seed 42
- 1Install golden-seed via pip
- 2Import UniversalQKD from gq and initialize a TestDataGenerator with seed=42
- 3Call random_user() to get a deterministic dict with id, age, and premium flag
- 4Re-initialize with seed=42 and call random_user() again
- 5Assert both dicts are identical to confirm full reproducibility
Two identical user dicts confirming the sequence is byte-for-byte reproducible from seed 42