The Agent Skills Directory

[COMMAND_EXECUTION]: The skill invokes the unity-mcp-cli tool to perform file reading operations on the host system.
[EXTERNAL_DOWNLOADS]: The instructions recommend installing unity-mcp-cli globally via npm (npm install -g unity-mcp-cli). This package is an external dependency that does not originate from a verified source or trusted organization.
[DATA_EXFILTRATION]: The filePath input parameter allows the agent to read arbitrary file contents. The skill lacks path validation or sandboxing, which could be exploited for path traversal to access sensitive information like .env files, SSH keys, or cloud credentials.
[INDIRECT_PROMPT_INJECTION]: The skill ingests data from local .cs files which could contain malicious instructions designed to manipulate the agent's behavior when the content is processed.
Ingestion points: The filePath parameter in SKILL.md defines how external file content enters the agent's context.
Boundary markers: None; there are no instructions or delimiters to isolate the script content from the agent's logic.
Capability inventory: The environment includes command execution via unity-mcp-cli and referenced write capabilities via script-update-or-create.
Sanitization: No validation or escaping is applied to the content read from disk.

script-read