The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests data from external repository files located in design/gdd/, design/narrative/, and design/levels/ to drive the level design workflow. This untrusted data is provided as context to subagents, creating a surface for potential indirect prompt injection.\n
Ingestion points: The skill reads game concept, pillars, narrative docs, and existing level documents from the local filesystem to provide context to subagents.\n
Boundary markers: The instructions do not define delimiters or specific warnings to ignore instructions embedded within the ingested files when passing them to subagents.\n
Capability inventory: The skill uses Task, Write, Edit, and Bash tools, allowing it to delegate complex tasks and modify repository content based on the processed design files.\n
Sanitization: The skill lacks sanitization or validation of the content read from the design documents before it is used to influence agent behavior.

team-level