The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to resolve and execute local CLI binaries (such as orca, orca-ide, or orca-dev) depending on the environment variables ORCA_CLI_COMMAND or ORCA_DEV_REPO_ROOT. These commands are used to interact with the accessibility tree, perform GUI actions (clicks, typing), and capture screenshots.
[DYNAMIC_EXECUTION]: The skill implements a 'discovery stub' architecture where it fetches its primary operational guide at runtime using the command ORCA skills get computer-use. This ensures the agent uses instructions that match the specific version of the installed Orca tool.
[INDIRECT_PROMPT_INJECTION]: By dynamically loading instructions from a command's output, the skill creates an ingestion surface where external data influences the agent's prompt context.
Ingestion points: The output of the ORCA skills get computer-use command is incorporated as a guide for the agent's task.
Boundary markers: None are explicitly defined in the stub to delimit the fetched guide from other instructions.
Capability inventory: The skill possesses capabilities for GUI manipulation including clicking, typing, and reading the accessibility tree via the orca computer toolset.
Sanitization: The skill relies on the integrity of the local orca binary to provide valid instructions.

computer-use