The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill requires downloading and executing external packages from the NPM registry, specifically @nomadamas/k-skill, k-skill-rhwp, and @rhwp/core.
[COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands using npx for tasks such as fetching dynamic instructions, listing files, and performing HWP document edits.
[INDIRECT_PROMPT_INJECTION]: The skill processes data from external .hwp and .hwpx files, which may contain untrusted content or embedded instructions.
Ingestion points: Document content is read into the agent's context through subcommands like info, search, and list-paragraphs as described in instruction.md.
Boundary markers: The instructions do not specify the use of delimiters or warnings to prevent the agent from interpreting data within the HWP files as instructions.
Capability inventory: The tool possesses powerful editing capabilities including insert-text, replace-all, and set-cell-text, which could be abused if the agent is misled by malicious document content.
Sanitization: There is no mention of sanitizing or filtering the content retrieved from documents before processing.

rhwp-edit