The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill describes workflows where an AI agent interacts with project source code and Roblox Studio via the Model Context Protocol (MCP). This creates a surface for indirect prompt injection if the code or data ingested by the agent contains adversarial instructions.
Ingestion points: Project source files and Roblox Studio DataModel accessed via MCP (documented in references/full.md).
Boundary markers: The instructions do not specify the use of delimiters or 'ignore embedded instructions' warnings for the data being processed.
Capability inventory: The skill facilitates the execution of various CLI tools (rojo, wally, selene, stylua) and file system modifications.
Sanitization: No sanitization or validation methods are proposed for external content before processing.
[EXTERNAL_DOWNLOADS]: The skill references multiple external documentation sources and community tools from domains such as github.com, rojo.space, wally.run, and create.roblox.com. These represent standard resources within the Roblox development ecosystem.
[COMMAND_EXECUTION]: The skill provides detailed instructions for executing command-line interface (CLI) tools associated with the Roblox toolchain, including environment setup, package management, and CI/CD operations.

roblox-tooling