The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes content from external sources when a user provides a URL for site migration. This creates a surface where malicious instructions embedded in the external site's content could influence the agent's behavior during the build process.
Ingestion points: User-supplied URL for migration or reference in Step 2, Round 1 ("Do they have an existing site to learn from or migrate? Get the URL").
Boundary markers: No specific delimiters or instructions to ignore embedded commands are defined for processing the external site data.
Capability inventory: The skill can write to the local file system (BLUTUI_BRIEF.md), execute courier CLI commands, and invoke MCP tools for site creation and configuration.
Sanitization: No explicit sanitization or validation of the external content is described.
[EXTERNAL_DOWNLOADS]: The instructions suggest adding a skill via npx skills add https://github.com/blutui/context. This involves fetching resources from the vendor's official GitHub repository to extend the agent's capabilities.
[COMMAND_EXECUTION]: The skill utilizes the courier CLI (e.g., courier push, courier mcp init) and Blutui MCP tools to perform site configuration, template building, and staging deployment. These operations are consistent with the skill's primary purpose of managing project lifecycles within the vendor's platform.

get-started