The Agent Skills Directory

[SAFE]: The skill consists of instructional markdown files and reference schemas designed to guide an agent through game balance analysis. It does not contain executable code, scripts, or commands that could compromise the host environment.
[SAFE]: No evidence of prompt injection or attempts to bypass AI safety guardrails was found. The instructions are focused on technical game design evaluation (e.g., exploratory ratios and human-limited simulation profiles).
[SAFE]: The skill does not perform network operations, access sensitive file system paths (like credentials or SSH keys), or include hardcoded secrets.
[SAFE]: There is no use of obfuscation, hidden content (zero-width characters/homoglyphs), or multi-layer encoding.
[SAFE]: The telemetry ingestion process described in references/log-contract.md and references/improvement-analysis.md is a structured data analysis task. While it involves processing external JSON logs, it does not involve executing the contents of those logs as code, mitigating typical injection risks.
[SAFE]: The skill follows least-privilege principles, focusing on analyzing simulation results and proposing structural code changes for game developers to review rather than executing arbitrary system commands.

evaluating-gameplay-balance