orca-cli
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to resolve and execute local binaries such as
orca,orca-dev, ororca-ide. The resolution logic depends on environment variables likeORCA_CLI_COMMANDandORCA_DEV_REPO_ROOT. - [INDIRECT_PROMPT_INJECTION]: The skill uses the command
ORCA skills get orca-clito dynamically fetch and load a version-matched guide into the agent's active context. This pattern allows external content to influence agent behavior at runtime. - Ingestion points: The output of the
ORCA skills get orca-clicommand specified inSKILL.md. - Boundary markers: The skill does not provide specific delimiters or instructions to the agent to treat the guide output as untrusted or to ignore embedded instructions within it.
- Capability inventory: The agent is granted capabilities to operate terminals, worktrees, and browsers via the
orcaCLI. - Sanitization: There is no evidence of content sanitization or validation for the data returned by the CLI before it is processed by the agent.
Audit Metadata