The Agent Skills Directory

[COMMAND_EXECUTION]: The skill facilitates the execution of the orca command-line utility and its environment-specific variants (orca-ide, orca-dev). These commands are used to interact with worktrees, terminals, and system processes.
[INDIRECT_PROMPT_INJECTION]: The skill implements a pattern where the agent is instructed to fetch its core operating instructions from the output of a local command (ORCA skills get orca-cli). This creates a surface where the tool's output could potentially include instructions that influence agent behavior.
Ingestion points: The output of the ORCA skills get orca-cli command as described in SKILL.md.
Boundary markers: Absent; the instructions direct the agent to load and follow the full guide directly from the command output.
Capability inventory: The skill provides access to terminal management (ORCA terminal list), worktree state (ORCA worktree ps), and browser control.
Sanitization: Absent; the agent is instructed to rely on the version-matched guide provided by the binary.
[DYNAMIC_EXECUTION]: The skill uses a dynamic approach to instruction loading by fetching its own manual at runtime from the orca executable, rather than relying on static documentation within the skill file.

orca-cli