The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local scripts bin/rmbc-upgrade and bin/rmbc-update-check, and sources lib/rmbc-common.sh from the skill's own directory structure. These are standard operations for an upgrade utility and are limited to the skill's known installation path.
[PROMPT_INJECTION]: The skill demonstrates an indirect prompt injection surface by ingesting and processing data from its environment and local files.
Ingestion points: Version strings parsed from preamble output and the contents of CHANGELOG.md (Step 3).
Boundary markers: Absent; the skill does not use delimiters to wrap the ingested content.
Capability inventory: Bash for command execution and Read for file access.
Sanitization: Absent; the changelog content is summarized directly for the agent. However, since the source is the skill's own package, the risk is consistent with standard application behavior.

rmbc-upgrade