The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to resolve and execute local binaries including orca, orca-ide, and orca-dev based on environment variables and session context.
[INDIRECT_PROMPT_INJECTION]: The skill uses a dynamic loading pattern where the actual usage guide is fetched from command output at runtime, creating a potential injection vector. \n- Ingestion points: The agent is instructed to run ORCA skills get orca-cli and treat the output as the authoritative guide for subsequent actions. \n- Boundary markers: No explicit boundary markers or instructions to disregard embedded commands in the output are provided. \n- Capability inventory: The skill grants the agent capabilities to manage file systems (worktrees), terminals, and browser sessions through shell commands. \n- Sanitization: There is no evidence of sanitization or structural validation performed on the retrieved content before it is incorporated into the agent's context.

orca-cli