The Agent Skills Directory

[SAFE]: The skill serves as a technical design framework for Reinforcement Learning in morphogenetic neural networks. It focuses on establishing safety governors, deterministic replay mechanisms, and fair evaluation baselines. The instructional content promotes rigorous engineering standards for AI training environments.
[SAFE]: No evidence of prompt injection or attempts to bypass agent safety guidelines was found. The instructions are purely technical and domain-specific, utilizing natural instructional language to guide the developer.
[SAFE]: No sensitive data access, credential exposure, or exfiltration patterns were identified. The included code snippets utilize standard PyTorch and distributed training primitives for illustrative purposes, such as RNG stream isolation and all-reduce operations for global statistics.
[SAFE]: The skill uses clear, unobfuscated language and standard mathematical constants for RNG seeding (e.g., the fractional part of the golden ratio). No hidden content, base64-encoded payloads, or malicious obfuscation techniques were detected.
[SAFE]: References to other skillpacks (e.g., 'yzmir-deep-rl', 'axiom-determinism-and-replay') represent a modular documentation structure within a consistent vendor ecosystem ('tachyon-beep') and do not involve untrusted remote code execution or unverifiable dependencies.

using-morphogenetic-rl