roblox-release-readiness
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external project data, repository state, and store metadata which provides a surface for indirect prompt injection. \n
- Ingestion points: The skill reads
AGENTS.md, release scope, design criteria, and external dashboard state (specified inSKILL.md). \n - Boundary markers: Audit activities are governed by a defined gate matrix (
references/gate-matrix.md) and a mandatory report format (references/readiness-report.md). \n - Capability inventory: The skill utilizes the
Roblox_StudioMCP tool for read-only inspection; no autonomous network exfiltration or file-writing capabilities are enabled in the instructions. \n - Sanitization: The instructions explicitly warn against converting tool failures into
PASSstatuses and require verifiable evidence for every observation. \n- [NO_CODE]: The skill consists of configuration files and markdown guidelines without any accompanying Python, JavaScript, or shell scripts, eliminating the risk of direct malicious code execution.
Audit Metadata