The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to fetch and follow instructions from the output of the ORCA skills get command. This creates a surface where instructions loaded at runtime could influence agent behavior beyond the static skill definition.
Ingestion points: The output of ORCA skills get orca-emulator-android is processed as instructions by the agent.
Boundary markers: No explicit delimiters or instructions to ignore embedded prompts are provided in the stub.
Capability inventory: The skill description indicates capabilities for adb control, application installation, logcat access, and hardware interaction.
Sanitization: No sanitization or validation of the fetched instructions is mentioned.
[COMMAND_EXECUTION]: The skill requires the execution of shell commands and local binaries (such as orca, orca-ide, or orca-dev) to function and fetch its core logic.
[DYNAMIC_EXECUTION]: The skill logic is not statically defined within the skill files; it is dynamically loaded at runtime from a version-matched source via the CLI. The path to the executable used is determined dynamically based on environment variables like ORCA_CLI_COMMAND and ORCA_DEV_REPO_ROOT.

orca-emulator-android