orca-cli
Pass
Audited by Gen Agent Trust Hub on Aug 9, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill facilitates interaction with the orca CLI tool. It provides logic for selecting the appropriate binary (e.g., orca-ide, orca-dev) and executing commands like status, open, and skills get.\n- [PROMPT_INJECTION]: The skill implements a dynamic instruction loading mechanism where the agent is told to run ORCA skills get orca-cli to obtain further operational guidelines. This creates a surface for indirect prompt injection if the output of the CLI tool were to be compromised or manipulated to include malicious instructions.\n
- Ingestion points: The output of the shell command ORCA skills get orca-cli is ingested into the agent context as instruction data (SKILL.md).\n
- Boundary markers: Absent; the agent is instructed to read and follow the output directly without explicit constraints on the content source.\n
- Capability inventory: The agent is granted shell command execution capabilities to interact with the Orca toolset, including worktrees, terminals, and internal browser controls.\n
- Sanitization: Absent; no sanitization or validation of the CLI output is performed before interpolation into the agent's prompt context.
Audit Metadata