orca-cli
Pass
Audited by Gen Agent Trust Hub on Sep 7, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill defines logic for resolving and executing local binaries (orca, orca-ide, or orca-dev) or a custom command path from the ORCA_CLI_COMMAND environment variable. This allows the agent to interact with the system's command-line interface.
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to fetch a version-matched guide using the ORCA skills get orca-cli command. This represents a vulnerability surface where external tool output is ingested directly into the agent's reasoning context. Ingestion points: The output from the ORCA skills get orca-cli command. Boundary markers: No delimiters or instructions are provided to the agent to treat the fetched content as potentially untrusted. Capability inventory: The agent can execute shell commands through the resolved binary. Sanitization: The skill lacks mechanisms to validate or sanitize the fetched guide before processing.
Audit Metadata