The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to fetch its full instruction set by running ORCA skills get orca-emulator-android and incorporating the output into its current context. This creates a surface for indirect prompt injection where the tool's output could potentially influence agent behavior beyond the intended scope.
Ingestion points: The agent is instructed in SKILL.md to run a CLI command and use the resulting output as a version-matched guide.
Boundary markers: Absent. There are no instructions to wrap the fetched content in delimiters or ignore embedded instructions within that content.
Capability inventory: The skill grants the agent the ability to execute shell commands via the orca CLI, including device control, app installation, and accessing accessibility trees.
Sanitization: Absent. The skill does not provide mechanism for the agent to validate or sanitize the instructions fetched from the binary.
[COMMAND_EXECUTION]: The skill defines logic for resolving the orca executable path using environment variables like ORCA_CLI_COMMAND and ORCA_DEV_REPO_ROOT, or fallback binary names like orca-ide. This logic is necessary for the skill's primary function of controlling an Android emulator.

orca-emulator-android