The Agent Skills Directory

[SAFE]: No malicious instructions, obfuscation, or unauthorized access patterns were detected within the skill's logic.
[PROMPT_INJECTION]: The skill features a context-gathering workflow that reads local project files (e.g., game concept, pillars, and narrative docs) to inform subagent prompts. This creates a surface for indirect prompt injection if project files are modified with malicious instructions, though the risk is minimal for local project assets.
Ingestion points: Project files in design/gdd/, design/levels/, design/narrative/, and design/art/.
Boundary markers: The skill does not define specific delimiters to isolate external content in prompts.
Capability inventory: Orchestrates subagents via Task; frontmatter allows Bash and Write tools.
Sanitization: No content validation or sanitization is performed on the ingested documentation.

team-level