The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill employs a pattern where instructions are retrieved dynamically from a local executable at runtime.
Ingestion points: The agent is instructed in SKILL.md to run ORCA skills get orca-cli and use the resulting output as the full usage guide.
Boundary markers: SKILL.md does not define specific delimiters or instructions to ignore potential commands embedded within the dynamic output.
Capability inventory: As described in SKILL.md, the skill possesses capabilities to operate worktrees, repositories, terminals, and an embedded browser through shell commands.
Sanitization: No sanitization or validation of the content returned by the binary is specified in SKILL.md before processing it as instructions.
[COMMAND_EXECUTION]: The skill is fundamentally designed to execute commands via the orca CLI to manage developer worktrees, terminals, and automation tasks.

orca-cli