The Agent Skills Directory

[PRIVILEGE_ESCALATION]: The skill explicitly instructs the agent to "re-run it with escalated permissions" if a runtime_access_denied error occurs. This encourages the agent to attempt to bypass security boundaries or sandbox restrictions imposed by the host environment.
[COMMAND_EXECUTION]: The skill facilitates the execution of shell commands using the orca CLI suite (including orca-dev and orca-ide). It provides logic for selecting the appropriate binary based on the environment.
[DYNAMIC_EXECUTION]: The skill instructions specify loading a version-matched guide at runtime using the command ORCA skills get orca-cli. This means the agent's behavior and available commands are partially defined by content retrieved dynamically from a local binary output.
[INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it loads instructions from external CLI output.
Ingestion points: The output of the ORCA skills get orca-cli command (SKILL.md).
Boundary markers: None identified; the retrieved content is processed directly as a guide.
Capability inventory: The skill has the capability to execute shell commands and interact with filesystems/terminals through the orca CLI.
Sanitization: None identified for the retrieved guide content.

orca-cli