orca-emulator-android
Warn
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: MEDIUMCOMMAND_EXECUTIONPRIVILEGE_ESCALATIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides logic for the agent to resolve and execute system binaries (
orca,orca-ide, ororca-dev). It also suggests running commands likeORCA openandORCA skills get. - [PRIVILEGE_ESCALATION]: The instructions explicitly tell the agent that if a command fails with
runtime_access_denied, it should "re-run it with escalated permissions." This encourages the agent to bypass security sandboxes using tools likesudoor administrative prompts. - [DYNAMIC_EXECUTION]: The skill uses a 'discovery stub' pattern where it fetches a "version-matched guide" at runtime using the
ORCA skills getcommand. This content then serves as the actual instructions for the agent to follow, representing dynamic instruction loading. - [INDIRECT_PROMPT_INJECTION]: By fetching external content (the guide) to drive agent behavior, the skill creates an attack surface for indirect prompt injection.
- Ingestion points: The output of the
ORCA skills get orca-emulator-androidcommand (SKILL.md). - Boundary markers: None provided in the stub to separate the fetched instructions from system context.
- Capability inventory: The skill possesses the capability to execute shell commands and interact with the local filesystem and
adbvia the Orca tool. - Sanitization: There is no mention of sanitization or validation of the fetched guide content before the agent processes it.
Audit Metadata