The Agent Skills Directory

[COMMAND_EXECUTION]: The skill requires the Bash tool to execute various game development commands, such as npx unity-mcp status. This provides the agent with the capability to run arbitrary shell commands within its environment as part of the intended workflow.- [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface (Category 8) because it reads and processes project-specific data to coordinate agent activities. Malicious content within these files could potentially influence agent behavior.
Ingestion points: The agent reads sprint-status.yaml, project-context.md, and various game design documents to manage workflows.
Boundary markers: No explicit markers are defined in the provided files to distinguish between trusted instructions and untrusted data within the project files.
Capability inventory: The skill has access to Bash command execution and filesystem Write tools, which could be leveraged if an injection occurs.
Sanitization: No specific mechanisms for sanitizing or validating the content of the processed files are described.- [EXTERNAL_DOWNLOADS]: The installation instructions utilize npx to fetch and add the skill from github.com/supercent-io/skills-template. This process downloads and executes external code from a third-party repository. It also references upstream documentation and module code from github.com/bmad-code-org.

bmad-gds