The Agent Skills Directory

[COMMAND_EXECUTION]: The skill is configured to execute specific commands via the agent-team CLI tool, including search, show, list, repo, and stats. This is a standard functional component for this type of skill.
[INDIRECT_PROMPT_INJECTION]: The skill processes and displays external content from a role catalog, which serves as an attack surface for instructions embedded in data.
Ingestion points: Content retrieved through the agent-team catalog search, agent-team catalog show, and agent-team catalog list commands (SKILL.md).
Boundary markers: The skill defines logical boundaries (e.g., read-only, browsing subset only) and requires reading a local rules file, but it does not provide explicit instructions for the agent to disregard natural language instructions found within the catalog metadata or role descriptions.
Capability inventory: Execution of the agent-team CLI for catalog metadata retrieval (SKILL.md).
Sanitization: There is no documentation of sanitization, filtering, or validation of the data retrieved from the catalog before it is ingested into the agent context.

catalog-browser