orca-emulator-android
Pass
Audited by Gen Agent Trust Hub on Nov 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONPRIVILEGE_ESCALATIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill uses a "discovery stub" pattern where the actual operational instructions are fetched at runtime using the
ORCA skills getcommand. This creates a surface where external content from the CLI tool can influence the agent's behavior. - Ingestion points: Output of the shell command
ORCA skills get orca-emulator-androidis loaded as a version-matched guide into the session context. - Boundary markers: Absent; the stub provides no delimiters or instructions to treat the fetched content as untrusted.
- Capability inventory: The skill is intended to control Android emulators via
adb, involving process execution, app installation, and hardware interaction. - Sanitization: Absent; the fetched instructions are interpolated directly into the context.
- [PRIVILEGE_ESCALATION]: The instructions advise the agent to "re-run it with escalated permissions" if a
runtime_access_deniederror is encountered. Instructing an agent to bypass sandbox restrictions via privilege elevation represents a security risk. - [COMMAND_EXECUTION]: The skill requires resolving and executing a local binary (
orca,orca-ide, ororca-dev) to function, which is the primary mechanism for its operations and configuration.
Audit Metadata