The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructions direct the agent to execute CLI commands to open generated HTML lesson files. This provides a mechanism for the agent to trigger local application behavior.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to fetch and incorporate information from external web resources to build 'Knowledge' within lessons. This creates a vulnerability where an attacker-controlled external source could contain malicious instructions designed to influence the agent or the user.
Ingestion points: External URLs and content sourced via RESOURCES.md and user-supplied mission parameters.
Boundary markers: No specific boundary markers or instructions to treat external data as untrusted are present in the skill instructions.
Capability inventory: The skill possesses the ability to write to the local file system (creating lessons, records, and assets) and execute shell commands to open files.
Sanitization: The instructions do not specify any sanitization or validation logic for data retrieved from external resources before it is processed or rendered.
[DYNAMIC_EXECUTION]: The skill workflow involves generating HTML files that include interactive components such as simulators and widgets. These components imply the generation and execution of scripts within the user's browser environment when the files are opened.

do-teach