The Agent Skills Directory

[PROMPT_INJECTION]: The skill instructions define scope and behavior without attempting to override safety guidelines or bypass agent constraints. Instructions remain focused on establishing functional game design rules.
[DATA_EXFILTRATION]: No evidence of hardcoded credentials, sensitive file path access, or unauthorized network operations was found. The skill operates on project-specific files as requested by the user.
[OBFUSCATION]: No Base64, zero-width characters, homoglyphs, or other encoding techniques were used to hide content. The content is written in plain, technical English.
[REMOTE_CODE_EXECUTION]: The skill does not download or execute external scripts from remote servers. It encourages the use of local project tooling and assets.
[INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied project briefs and writes game design artifacts to the project directory.
Ingestion points: Text following the /worldview-game-threat-interrupted-puzzle command as described in SKILL.md.
Boundary markers: None specific to the skill body, relying on standard agent delimiters and project scoping instructions.
Capability inventory: File read/write operations for documentation (mechanic.md, tunables.yaml, verification.md) and implementation logic within the project directory.
Sanitization: None specified for input briefs.
Assessment: This ingestion surface is a core component of the skill's primary function as a design assistant and does not represent an elevated security risk.

worldview-game-threat-interrupted-puzzle