The Agent Skills Directory

[SAFE]: The skill operates on local game design documents within the design/gdd/ and production/ directories. No network requests, external data exfiltration, or access to sensitive system credentials (e.g., .ssh, .aws) were detected.
[SAFE]: A collaborative design protocol is strictly enforced, utilizing AskUserQuestion to obtain user confirmation before every major step, including system enumeration, dependency mapping, and final file writes.
[SAFE]: The skill uses the Task tool for internal agent coordination (spawning technical, producer, and creative director reviews), which is a standard pattern for multi-agent workflows and does not involve remote code execution from untrusted sources.
[SAFE]: The skill processes local project files to generate a systems index, representing a standard documentation workflow.
Ingestion points: Reads game concept and pillar documents from the local design/gdd/ directory.
Boundary markers: None explicitly implemented; the skill relies on model processing and user review gates.
Capability inventory: File read/write/edit and internal task spawning.
Sanitization: The skill uses model-driven decomposition rather than raw interpolation, providing a layer of natural language processing between input data and file output.

map-systems