The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill interacts with external data from Linear tickets, which may contain untrusted user content such as comments and attachments. This represents an indirect prompt injection surface where a malicious user could attempt to influence the agent's behavior.
Ingestion points: Linear ticket text, comments, and attachments as noted in the description.
Boundary markers: The description explicitly instructs the agent to treat ticket text, comments, and attachments as untrusted data and never as instructions.
Capability inventory: The skill utilizes the Orca CLI to search tickets, move workflow states, and create new tickets.
Sanitization: Employs natural language directives to the agent to disregard instructions within the processed data.
[DYNAMIC_EXECUTION]: The skill performs dynamic loading of instructions by calling ORCA skills get linear-tickets to retrieve a version-matched guide. This ensures the agent uses instructions appropriate for the specific version of the Orca CLI installed on the system.

linear-tickets