The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests data from external sources that may contain instructions.
Ingestion points: Data is pulled from external upgrade guides via kirby:kirby_online and kirby:kirby_search, and from local pages via kirby:kirby_render_page.
Boundary markers: There are no delimiters or instructions provided to the agent to treat fetched content as data rather than instructions.
Capability inventory: The skill utilizes tools like kirby:kirby_runtime_install and executes project scripts found during audits, providing a path for execution of influenced logic.
Sanitization: There is no process described for sanitizing or validating the content retrieved from external or local sources before processing.
[COMMAND_EXECUTION]: The skill facilitates the execution of system commands and project-specific scripts through tools like kirby:kirby_runtime_install and the processing of composer audit results. While intended for maintenance, these capabilities could be misused if the agent is influenced by malicious input.

kirby-upgrade-and-maintenance