The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests data from external sources including backlog items, glossary terms, and project files, which constitutes an indirect prompt injection surface.\n
Ingestion points: Context is loaded from external backends via rota item show <ID>, rota glossary read <term>, and the protocols in references/context-load-protocol.md.\n
Boundary markers: The skill mentions DECISIONS comments as hard boundaries for conflicts but does not specify robust delimiters for general ingested text.\n
Capability inventory: The skill can execute commands that write to the project backend, specifically rota design add, rota design put, and rota item comment add.\n
Sanitization: No sanitization or escaping protocols are defined for the data retrieved from external sources before it is used in design drafts.\n- [COMMAND_EXECUTION]: The skill utilizes a series of CLI commands through the rota tool to manage the design process. These commands interact with the project's backend to resolve items, load details, and store approved documentation.

rota-brainstorm